A fatal shipping accident in Brisbane demonstrates the dangers of complacency, even when undertaking routine tasks.

On 15 December 2011, the crew of the British Beech were taking on stores from a barge when, during the return of a stores container from the ship to the barge, the container came free of its slings. The container fell to the barge below, striking the master of the barge. The barge crew and shore paramedics attended to the master, but he died from his injuries while being transported to hospital.

The barge master placed himself in a position of danger under the suspended load, and the barge crew had not followed their company procedures for storing operations.

Lifting operations, even when they are routine, involve inherent risks. The ATSB investigation found that the ship’s crew did not view the storing operation as dangerous and had, over time, removed identified safety barriers which would probably have prevented the accident. The container had not been appropriately rigged on board the ship and the ship’s crew had not warned the barge crew of its return.

In addition, on the barge, the barge master placed himself in a position of danger under the suspended load, and the barge crew had not followed their company procedures for storing operations.

The companies of both vessels were found not to have adequately implemented compliance auditing processes. In the case of the barge company, there had been previous incidents which, though less serious, had not been acted on or learnt from. Effective compliance auditing could have caught and pre-empted some of the dangerous practices which resulted in the accident. The ATSB urges that established procedures be followed, reinforced and audited in order to ensure that vigilance is maintained and complacency avoided.

As a result of the accident and investigation, the method for handling these types of containers in Brisbane has changed so that the containers are top lifted and slings are no longer used. The ship’s manager, BP Shipping, has also implemented a thorough review of lifting and slinging processes, practices, procedures and equipment. A fleetwide review and training workshop for lifting and slinging was also completed.

This is one of several accidents that the ATSB has investigated which involved unsafe working practices in the maritime industry. Unsafe maritime work practices forms the subject of one of the ATSB’s Safety Watch articles. Safety Watch identifies safety concerns that deserve heightened attention.

Full details of the safety actions taken by both the ship’s manager and the barge’s operator can be found in the investigation report: Fatality while storing the products, tanker British Beech, in Brisbane, Queensland, on 15 December 2011

The ditching of a Robinson R44 helicopter into the ocean 80 km north of Horn Island shows how good preparation greatly improves the chances of survival and rescue.  The ditching took place on 9 June 2012 during a return flight from Dauan Island to Thursday Island.

Earlier that day, on the way to Dauan Island, the helicopter’s alternator light illuminated on two separate occasions. Both times, the pilot turned the alternator off and then back on and the light went out. Later, when the pilot started the engine for the return flight from Dauan Island, the engine rotated several times and made a clicking sound ‘like a battery without enough power’.

After consulting with the operator, the pilot used truck batteries to start the helicopter. He ran the engine at idle power for about 10 minutes before departing for Horn Island.

Fortunately, the pilot had taken precautions before commencing the flight—precautions that enabled him to land the helicopter safely, summon help, and be located by the search and rescue team.

About 10 minutes after departing, the alternator light illuminated again. The pilot turned the alternator off and back on again and the light went out. This happened three more times before the pilot isolated all non-essential electrical systems. By then, he had passed the point of no return to Dauan Island and decided to fly to Moa Island instead. Continued trouble with the engine, however, indicated a possible engine overspeed, so he deployed the helicopter’s emergency ‘pop-out’ floats and landed in one metre of swell. The pop-out floats allow the helicopter to float on the water surface.

Fortunately, the pilot had taken precautions before commencing the flight—precautions that enabled him to land the helicopter safely, summon help, and be located by the search and rescue team.

A few years previously, he had undertaken ditching training, and was prepared for a water landing. After landing safely, he shut down the helicopter and, since the conditions were reasonably stable, remained in the cockpit, wearing a life jacket equipped with flares. He activated his personal locator beacon (PLB) and attempted to contact the Horn Island police on his mobile phone, but the signal dropped out. He then used his mobile phone to contact the helicopter operator, who initiated a search and rescue operation by contacting the Rescue Coordination Centre Australia. Shortly after, the signal from the PLB was detected by the Cospas-Sarsat satellite system. In addition, the pre-arranged SARTIME (a time nominated by the pilot for search and rescue proceedings to begin) was reached, and the automated software initiated.

The crew of the search and rescue helicopter were not able to determine an accurate location from the PLB signal since it remained within the helicopter fuselage, but they navigated to the pilot using the flares that he discharged. The pilot was rescued uninjured; however, the helicopter sustained substantial damage due to the salt water. It was recovered the next day.

A detailed examination of the helicopter following the accident found evidence of an engine overspeed. As a result of this occurrence, the aircraft operator has, among other actions, amended company policy in relation to a flat battery and faulty alternator, requiring immediate replacement. In addition, the lanyard on the PLB has been extended from 0.5 m to 3 m, allowing the PLB to be thrown clear of the helicopter. Pilots have also been briefed in the use of noise-cancelling headsets, after the pilot reported that his headset may have dampened the abnormal engine sounds.

The full report of AO-2012-096 is contained within the Aviation Short Investigation Bulletin – Issue 15.

An incident at Jabiru, Northern Territory demonstrates the importance of both un-alerted and alerted see-and-avoid principles and of pilots checking the serviceability and correct operation of radio equipment before flying.

On 5 October 2012, a Beech 1900, registered VH-EMK (EMK) departed Darwin on a charter passenger flight to Jabiru. When descending into Jabiru, the crew broadcast on the Brisbane Centre frequency advising that they were 40 NM west of Jabiru, and were due to arrive at 0654. Brisbane Centre air traffic control advised that there was no traffic for the descent.

At about 0645, an Airparts FU-24 aircraft, registered VH-HVP (HVP) was being prepared for an aerial survey flight in the Jabiru area. During flight preparations the pilot turned the aircraft’s radio on and selected standby on the transponder. Shortly after, the pilot broadcast on the common traffic advisory frequency (CTAF) that he was taxiing for runway 27. At that time the pilot heard a broadcast from the crew of EMK advising that they would be established on a 5 NM final at 0654. The pilot of HVP determined that he would have 4 minutes to depart before EMK was reported to be established on its final approach.

Pilots should make use of all available resources such as an Aerodrome Frequency Response Unit (AFRU) to confirm radio serviceability.

The pilot of HVP broadcast a call advising that he was entering and backtracking runway 27 and selected ‘ALT’ on the transponder. The pilot reported that he received no reply to his broadcast and believed there was no conflict with EMK.

When about 1 NM inbound, the crew of EMK saw HVP taking off on runway 27 directly opposite their approach path. The captain immediately called for a go-around, which the first officer (FO) initiated. The FO took avoiding action by manoeuvring to the right. At the same time the pilot of HVP saw EMK on final for runway 09 and decided to continue the take-off as he did not wish to remain on the runway if the crew of EMK had not seen HVP. After take-off, the pilot of HVP turned the aircraft to the right to maintain separation from EMK.

Following the incident the crew of EMK tried to contact HVP three times, but received no reply and HVP was not observed on the TCAS (traffic collision avoidance system) display. The pilot of HVK heard a broadcast from EMK and tried to respond, but realised that his radio was only receiving and not transmitting. The two crews reported different assessments of the minimum separation of the two aircraft, but it appears that separation reduced to about 300 ft vertically and 200 to 250 m horizontally.

HVP’s communications equipment comprised a VHF radio, a HF radio and a satellite phone. A single radio selector switch was used to activate each system. The pilot of HVP reported using the satellite phone the previous day. When changing the selection back to VHF he had inadvertently placed the selector between the VHF and HF settings. He did not confirm the radio selection during pre-flight checks. The crew of EMK reported that HVP was not observed on the TCAS.

The practice of ‘see-and-avoid’ has long been the primary method for minimising the risk of collision when flying in visual meteorological conditions in uncontrolled environments and is considered a crucial element of a pilot’s situation awareness. The use of a radio, combined with a visual lookout, markedly increases effectiveness; however, pilots need to be mindful that the absence of a traffic broadcast does not necessarily mean the absence of traffic. In addition, pilots should make use of all available resources such as an Aerodrome Frequency Response Unit (AFRU) to confirm radio serviceability.

Read the final report: Aircraft proximity event between Beech 1900, VH-EMK, and Airparts FU-24, VH-HVP, Jabiru Airport, Northern Territory, on 5 October 2012

Limitation of the see-and-avoid principles

The Australian Transport Safety Bureau urges pilots to take the time to ensure they have all the information they need before landing their aircraft. This warning comes after an accident in South Australia where a Piper PA-39 skidded off the end of the runway at Innamincka, before spinning to the left and coming to rest in a gully.

The aircraft was carrying the pilot and one passenger. The landing area at Innamincka had one gravel runway, about 1,000 m long. As the aircraft approached the airfield, the pilot noticed that the windsock was indicating a strong crosswind. As a precaution, he increased his approach speed and reduced his flap setting. The pilot elected to land further along the runway than normal to avoid the rough, rocky ground just before the runway. As a result, the aircraft touched down about a quarter of the way along the runway.

Pilots should establish a decision point along the runway at which a go-around should be initiated if the requirements for a safe landing can no longer be met.

When the aircraft was about halfway along the runway, the pilot realised that it was going too fast, so he applied full braking. He judged it was too late to commence a go-around, but the braking was ineffective due to the surface of the runway. The aircraft continued on beyond the end of the runway before its left wheel struck a depression. This caused it to spin to the left before coming to rest in a one-metre-deep gully. The pilot sustained minor injuries and the passenger was uninjured, but the aircraft was significantly damaged. The pilot recalled the surface of the runway contained a lot of small loose stones, which may have affected the braking capacity of the aircraft during the landing.

This accident demonstrates the importance of assessing the operational and environmental conditions at the time to determine the most suitable landing type. Pilots should also establish a decision point along the runway at which a go-around should be initiated if the requirements for a safe landing can no longer be met.

The accident also highlights the benefits of using all available resources, including people on the ground, for gathering information on the actual conditions.

Read the final report: Runway excursion involving Piper PA-39, VH-MMN, Innamincka Township (ALA), South Australia, on 26 October 2012, which contains links to publications on short field approaches and landings.

You can now get the latest ATSB news and updates via our Twitter account: @ATSBinfo(Opens in a new tab/window).

We will use Twitter to tell you about a range of activities and initiatives, including:

• new and updated investigations
• investigation and research report releases
• new safety awareness products.

Some things to keep in mind
The ATSB will not guarantee responding to tweets.

If you have a general enquiry, you should contact the ATSB via email atsbinfo@atsb.gov.au or
phone 1800 020 616.

You should not use Twitter to report a transport safety occurrence. To make a mandatory accident and incident notification:

• call 1800 011 034
• submit an online form on the notifications page of the ATSB website
• fax 02 6274 6434.

On 17 November 2012, a pilot approaching Geelong (Grovedale) airstrip in a Piper PA-28R was feeling some apprehension: he did not hear any broadcasts from aircraft operating at the airstrip and could not see any aircraft on the ground.

The pilot considered diverting to Barwon Heads, but for a number of reasons, he decided to continue to Geelong. He conducted a precautionary flight over the runway, and he and his passenger saw no signs indicating the airstrip was closed. There were cars and umbrellas on the ground, and the runway appeared the same as it had when he had visited on previous occasions. He elected to land.

The pilot had a message on his home phone and mobile from Airservices Australia
advising that Geelong was closed.

After landing safely, the pilot noticed that the office buildings were unoccupied and a fence had been placed across one of the runways. He was later told by people on the ground that the airstrip had been closed since April and was going to be redeveloped as a residential estate.

This incident demonstrated the vital importance of reviewing all available flight information, including the condition and suitability of the selected landing areas. When preparing for the flight, the pilot looked at a number of potential landing areas including Barwon Heads and Geelong. The pilot initially referenced the En Route Supplement Australia (ERSA) and noted that there was airstrip information for Barwon Heads, but not Geelong. He then referred to the Aircraft Owners and Pilots Association of Australia (AOPA) National Airfield Directory 2010/11 to obtain runway information for both airstrips. The latest edition of the AOPA Directory (2012) was released at about the same time the incident occurred. It stated that the Geelong (Grovedale) airstrip was closed.

When an aerodrome is rendered completely unserviceable for all operations, an unserviceability cross marker is displayed in the signal circle (a coloured area near the windsock that is used for displaying ground signals to pilots). During the precautionary flyover, the pilot and passenger did not see any cross markers, although they did observe a light coloured section near the end of the runway. 

After landing, the pilot contacted Airservices Australia to cancel his SARTIME (the time nominated by a pilot for the initiation of Search and Rescue action if a report from the pilot has not been received by the nominated unit.) As they acknowledged the cancellation, the Airservices employee made reference to Barwon Heads as the planned destination. When he returned to Bairnsdale, the pilot noted that he had a message on his home phone and mobile from Airservices Australia advising that Geelong was closed and his flight planned destination had been changed to Barwon Heads.

Read the final report: Landing on a closed airstrip involving a Piper PA-28R, VH-HKZ, Geelong (Grovedale), Victoria, on 17 November 2012, which provides more detail on the incident and the advice of the ATSB.

Pilots and operators of Bombardier DHC-8-100, -200 and -300 series aircraft are being urged to take steps to reduce the risk of propeller overspeed following an incident involving a DHC-8-315 aircraft. 

The incident occurred on 6 December 2011 when the aircraft, operated by QantasLink, was on a scheduled flight from Cairns to Weipa in Queensland. The first officer was the pilot flying. The aircraft was descending, with the power levers in the flight idle position and the first officer’s hand on the power levers, when it entered a layer of cumulus cloud. There were a few small bumps from the turbulence. A few seconds later, the aircraft encountered a strong updraft followed by a downdraft, which caused the first officer to inadvertently lift one or both of the idle gate release triggers and move the power levers below the flight idle gate 

The flight idle gate is a mechanical stop that assists with preventing the power levers from going into ground beta mode. Ground beta is a control mode used to control the pitch of the propeller blades during ground operations. Because ground beta mode inhibits propeller speed governing, pilots are prohibited from using it while in the air. 

The ATSB found that while the beta warning horn sounded as designed, the pilots were not acutely aware of the purpose of the warning horn due to a lack of previous exposure to the sound.

In the cockpit, the beta warning sounded, and the propeller noise increased. The captain initially thought that the warning horn was the sound of the autopilot disengaging, but realised that was not the case as soon as he heard the propeller noise increase. The first officer realised the situation, and immediately moved the power levers forwards of the flight idle gate.

During the short time that the aircraft’s power levers were in the ground beta range, both propeller speeds increased uncontrollably by over 300 revolutions per minute. 

The crew found no abnormal engine indications as a result of the occurrence and the flight proceeded normally to Weipa. Subsequent maintenance checks after the flight did not reveal any engine or propeller system damage.

The aircraft design included features to reduce the likelihood of the power levers being moved into the ground beta mode during flight. However, the ATSB found that many DHC‑8-100, -200 and -300 series aircraft did not have any way of preventing movement of the power levers below the flight idle gate in flight, or a means to prevent such movement resulting in a loss of propeller speed control. This design limitation has been associated with several safety occurrences. 

As of March 2012, there were 57 DHC-8-100, -200 and -300 series aircraft registered in Australia. Some of the aircraft were acquired second-hand from the US and had protective measures previously installed as required by US airworthiness requirements. The ATSB contacted the DHC-8 operators and determined that, at the time of this occurrence, there were 52 aircraft in Australia that were not protected against the consequences of a flight crew moving the power levers below flight idle in flight.

The aircraft manufacturer has advised that it will be releasing a Service Bulletin modification to fix the propeller speed control issue. That bulletin will be mandated by an Airworthiness Directive from the airworthiness authority of the State of Design (Canada) to ensure that the bulletin is incorporated into all the aircraft affected by the design issue worldwide, including those in Australia. In addition, the aircraft operator has introduced a series of actions to reduce the risk of such occurrences. 

The ATSB found that while the beta warning horn sounded as designed, the pilots were not acutely aware of the purpose of the warning horn due to a lack of previous exposure to the sound. In response, the ATSB has released an extract from the cockpit voice recorder with the beta warning horn sounding and the audible rise in propeller speed evident to all Australian operators of the aircraft type. The audio file is also available on the ATSB website to help increase awareness of the issue.

Read the final report: Double propeller overspeed involving Bombardier DHC-8, VH-SBV, near Weipa, Queensland, on 6 December 2011, which provides the full report and the sample of the cockpit sounds.

The ATSB investigation into a serious incident at Melbourne Airport, Victoria has demonstrated the requirements and complexities that come with the use of automatic flight systems. While modern automation can greatly reduce flight crew workload, crews still need to maintain a thorough understanding of those systems. 

The incident occurred on 24 July 2011 when, at 2019 EST, a Thai Airways Boeing 777-3D7 aircraft was conducting a runway approach to Melbourne Airport after a flight from Bangkok. The approach was being made using very high frequency omnidirectional radio range (VOR) — a ground-based navigation aid that emits a signal that can be received by appropriately‑equipped aircraft and represented as the aircraft’s bearing to or from that aid.  

The ATSB established that the pilot in command may not have fully understood some aspects of the aircraft’s automated flight control systems and probably experienced ‘automation surprise’...

During the approach, the tower controller observed that the aircraft was lower than required and asked the flight crew to check their altitude. The tower controller subsequently instructed the crew to conduct a go-around. However, while the crew did arrest the aircraft’s descent, there was a delay of about 50 seconds before they initiated the go-around and commenced a climb to the required altitude.

The ATSB established that the pilot in command may not have fully understood some aspects of the aircraft’s automated flight control systems and probably experienced ‘automation surprise’ when the aircraft pitched up to capture the VOR approach path. Automation surprise occurs when an automatic system changes its operation unexpectedly. 

As a result, the remainder of the approach was conducted using the autopilot’s flight level change mode. In that mode, the aircraft’s rate of descent is unrestricted and therefore may be significantly higher than that required for an instrument approach. In addition, the flight crew inadvertently selected a lower than stipulated descent altitude, which resulted in descent below the specified segment minimum safe altitude for that stage of the approach and the approach not being managed in accordance with the prescribed procedure.

In response to this occurrence, Thai Airways International issued a notice to flight crews that emphasized the importance of constant angle non-precision approaches and adherence to the segment minimum safe altitudes. Other actions included a review of crew training in support of non-precision approaches and the provision of additional information relating to the use of the aircraft’s autopilot flight director system.

This occurrence highlights the risks inherent in the conduct of non-precision approaches and reinforces the need for flight crews to closely monitor the aircraft’s flight path to ensure it complies with the prescribed procedure. 

Modern air transport aircraft are equipped with ever increasing levels of automation. While flight crews retain the option of flying the aircraft manually, the use of automation is generally preferred and often provides increased levels of safety and efficiency. Worldwide, errors associated with the use and management of automatic flight systems have been identified as causal factors in more than 20 per cent of approach and landing accidents.

Read the final report: Operational non-compliance involving Boeing 777-3D7, HS-TKD, 15 km south of Melbourne Airport, Victoria, on 24 July 2011, which provides more detail on the incident and the measures taken to remedy it.

An ATSB investigation has uncovered a 27 year-old omission that rendered an aircraft manufacturer’s safety bulletin ineffective. As a result of this discovery, the manufacturer issued a new bulletin to address the issue, which was then made mandatory by Transport Canada and the Civil Aviation Safety Authority of Australia.

On 30 December 2011, a Bombardier DHC-8-102 was being operated on a scheduled passenger service to Cairns, Queensland. During the landing, it seemed to the crew that the aircraft decelerated much more quickly than they expected, given that reverse thrust and landing gear brakes had not been selected.

An inspection of the aircraft found nothing to explain the perceived problem, and a review of the flight recorder data indicated that there had been no abnormal operation of the engines or propellers, and that reverse thrust had not been used. There was little if any risk associated with the event. However, a subsequent inspection of the aircraft discovered a design problem with the aircraft’s power lever controls. 

Once informed of the design problem,
the aircraft manufacturer took prompt action
to address the issue.

The problem related to the friction device within the aircraft’s power levers control quadrant and its interaction with the flight idle gate, which was itself designed to prevent the power levers from going into the ground range while in flight. Operation of the power levers in the ground range slows the aircraft after landing by changing the pitch of the propellers to create aerodynamic drag loads and providing a considerable amount of reverse thrust. The problem with the friction device meant that with power lever friction selected fully off, the flight idle gate could be rendered inoperative.

The design problem only applied to the first 39 DHC-8-100 aircraft that were manufactured; subsequent aircraft were manufactured with a modified design. In 1986, the aircraft manufacturer had introduced a service bulletin requirement to modify these 39 aircraft retrospectively, but the service bulletin omitted a requirement to modify or replace a specific part, which resulted in the bulletin being ineffective.

Once informed of the design problem, the aircraft manufacturer took prompt action to address the issue. They issued a service bulletin to modify the relevant part. 

The results of this investigation show how important it is that crews report occurrences and other perceived problems. Although in this case the actual event reported by the crew was not serious, and no problems relating to the aircraft or crew performance leading to the perceived event were identified, the subsequent investigation did identify a safety issue in the design of the aircraft.

More details of the investigation, along with the findings, can be found in the investigation report AO-2012-005

The ATSB’s Chief Commissioner, Martin Dolan, has just published his latest blog post on unsafe work practises.

In his post, Martin talks about a fatal accident involving a seaman who was knocked off a ladder by a wave while working over the side of a container ship. The lessons we’ve learnt from this and other accidents aren’t just limited to the maritime industry. They apply to all operations that deal with risk.