Break-up

In accordance with the flightplan lodged before departure, the aircraft tracked direct to Moree and on to Coonamble, cruising at 10,000ft. But at 1. 03pm, it amended its destination to Swan Hill, Victoria. Ten minutes later the aircraft reported over Coonamble, maintaining 10,000ft, and estimating Hillston at 2. 18pm.

At 1. 16pm, it called again, reporting it was diverting five nautical miles left of track because of weather, and 20 minutes later further advised it would have to divert up to 20nm left of track. Another call at 1. 48pm extended this diversion to 29nm left of track.

This proved to be the last transmission from the Chieftain. At about 2pm, police at Condobolin received a telephoned report that an aircraft had crashed on a farming property about 28km north of the town.

Extensively burnt-out main wreckage was subsequently confirmed as that of VH-PYN.

Other wreckage was found the next day, spread along a trail up to four kilometres from the main wreckage.

Investigation

Recorded ATC radar data showed the aircraft entering radar coverage about 50km north of Condobolin just after 1. 46pm, and its last radar return was recorded three minutes and 19 seconds later. During this brief period, the aircraft’s track was about 56km left of the direct Coonamble-Hillston track and it changed direction from southerly to southwesterly. Its ground speed was between 200 and 220kt and its altitude remained at

The rear fuselage and empennage showed evidence of severe torsional twisting to the right, typically incurred by an extreme rate of roll in the opposite direction.

10,000ft. The last observed radar position was close to the limit of radar coverage and within 10km of where its wreckage was found.

Wreckage

The aircraft had struck the ground upside down at low forward speed, and was destroyed by a fuel-fed fire. Sections of the control surfaces, together with the starboard engine, had separated in flight, and both wings had failed outboard of the engine nacelles. The starboard engine was lying about 300m from the fuselage. The rear fuselage and empennage showed evidence of severe torsional twisting to the right, typically incurred by an extreme rate of roll in the opposite direction.

Damage to the engines and propellers was consistent with the effects of the inflight break-up, impact, and fire. The port propeller, still attached to its engine and the aircraft, was not rotating at impact. One blade had broken off, having failed in overload, possibly when struck by a part of the aircraft during the break-up. The separated starboard engine was not fire-damaged and its propeller blades were feathered, but there was evidence that it was rotating when the engine was torn from the wing.

Numerous pieces of the outer wings, tailplane and nose were found along the wreckage trail. A section of the starboard wing leading edge was torn off in a forward direction, implying the wing was not moving forward when the section separated. But deformation to port wing skin pieces suggested the skin peeled off rearwards. The starboard aileron was bent 60 degrees upwards at about its mid-span, having been forced in a spanwise direction. Although the aileron trim tab and its control rod remained attached, both trim cables had failed in overload. On the port aileron, a mid-span tear indicated it had folded upwards through at least 90 degrees during the break-up. There was no evidence of repetitive, hard contacts on any of the aileron stops.

Damage to the tailplane suggested it was struck by part of a wing or the starboard engine. The elevator trim tab control rod remained attached, but both control cables had failed in overload. The threaded shaft of the screw jack mechanism had also failed in overload.

The Aircraft

Manufactured in 1982, the Chieftain had flown a total time of only 2900 hours. It had recently been fitted with new engines, propellers, and fuel tanks, and was equipped with a King KFC 200 autopilot and a Garmin GNS530 GPS unit coupled to the autopilot.

The Pilots

The pilot-in-command held a commercial licence, a command multi-engine instrument rating and a Grade 1 instructor rating. His total flying experience was around 4600 hours, more than 1000 hours of which was in PA-31 aircraft extending over 13 years. Flying VH-PYN had been practically his full-time occupation for the previous 20 months.

The other pilot was also employed by the company that operated VH-PYN. He held a commercial licence and a command multiengine instrument rating. His total flying experience was about 1560 hours, and he had recently been endorsed on the PA-31, gaining about six hours experience on the type. He had a good working relationship with the pilot- in-command, whom he had known since 2001. Both pilots had experience in unusual attitude recovery, had undergone training for low-level powerline patrol operations, and had some experience of aerobatics.

The aircraft owner was also a pilot who had flown VH-PYN for some years. It was his policy that the aircraft, while fully equipped for IFR, should be flown only in visual conditions. Said to have been a “fussy flyer”, he was averse to taking risks to save time or money, particularly in regard to weather.

The pilot-in-command enjoyed a close working relationship with him, and was familiar with his requirements. There were numerous occasions on which they had cancelled or postponed trips because of weather.

It was the owner’s policy that the aircraft, while fully equipped for IFR, should be flown only in visual conditions.

Meteorological Forecasts

The pilot-in-command had obtained a weather forecast before leaving Archerfield. The Area 22 forecast, issued at 2. 20am that morning indicated thunderstorms, associated with a trough moving through central NSW, east of a line from Bourke to Griffith, after 10am. A new Area 22 forecast, issued at 9. 30am, contained no significant changes, but at 11. 30am it was amended to indicate thunderstorm activity had increased from ‘tending occasional’ to ‘frequently observed’ east of a line from Bourke to Griffith. It included reference to a SIGMET reporting frequent thunderstorms within 60nm of a line from Cobar to Wagga. Airservices received this SIGMET five minutes after the aircraft left Archerfield.

Actual Weather

An active front was moving east at 15 to 25kt through New South Wales. Satellite images showed a line of thunderstorms extending from south-central Queensland, through New South Wales, and into Victoria. A line of active thunderstorms marked the front’s two forks. Weather radar showed a line of precipitation passing through Condobolin around 1. 50pm, with an almost continuous line of storms extending south-south-west to northnorth- east, from just northwest of Wagga. The line formed a ‘Y’ just north of Condobolin near the site of the accident. Later, when the storms were in better radar range, the storm complex appeared to extend unbroken for some distance to the north.

Radar images showed individual storm cells close to the aircraft’s position at the time of the accident.

Immediately beforehand, VH-PYN would probably have been surrounded to the west, south and east (ie inside the ‘Y’) by a large complex of storms.

Atc Action

At 1. 17pm, shortly after the aircraft reported over Coonamble, VH-PYN was instructed to call the Bogan airspace Sector Controller. The controller responsible assessed that the SIGMET did not meet requirements for alerting VH-PYN – thunderstorms were included in the Area 22 forecasts, and the aircraft was still some distance from the hazardous area.

More than two hours before at 11. 5am, the Sector Controller had issued an “all stations” hazard alert because of an amended Nyngan Terminal Area Forecast that included a TEMPO period. He repeated it at 12. 15pm and 12. 48pm when requests came in from aircraft tracking to Bourke and Parkes for diversions because of the weather.

His 12. 48pm broadcast included changes to forecasts for Condobolin and Cobar, the controller advising pilots to contact FLIGHTWATCH for more information.

At 1pm, the controller checked whether the pilot of another Piper Chieftain, en route from Bourke to Cobar, had received the amended terminal area forecast for Cobar. Between 1. 20 and 1. 25pm, the controller received further requests for diversions from a Saab 340, 11nm southeast of Griffith; a third Piper Chieftain tracking to the south-east of Bourke; a Boeing 737, tracking south from the FIR boundary; and an A320 on a similar track to the 737.

At 1. 25pm, because of increasing demands the weather situation was making, the sector workload was split between two controllers, and between 1. 25 and 1. 35pm there were two more requests for diversions.

But no transmissions came from VH-PYN about hazardous weather, and at no time during its flight did it ask for a weather update. It was possible of course that its crew overheard FLIGHTWATCH transmissions to other aircraft on the weather situation.

Analysis

The dispersion of the wreckage left no doubt that the aircraft broke up in the air. Yet there was nothing to suggest that the flight was other than normal before radar contact was lost.

Furthermore, the wreckage examination revealed no pre-existing condition that could have led to the aircraft breaking up at less than its design load limit. The absence of any emergency transmission suggested some sudden, major catastrophe, with little or no warning, denying the pilots of any chance to transmit a distress call.

None of the recovered wing panel and aileron sections, the elevators, or the elevator trim tab exhibited evidenceof deformation from aerodynamic flutter, and it was unlikely that flutter contributed to the break-up.

Damage to the outer wings indicated loads great enough to deform the structure, resulting in lifting and loss of wing skin. Aerodynamic forces could then have progressively shed wing pieces, resulting in loss of structural integrity and destabilising the aircraft in yaw, pitch, and roll. The twisting of the empennage showed the aircraft was subject to extreme rolling forces during the break-up.

There was no evidence that the aircraft had been struck by lightning or of any in-flight explosion such as a fuel tank exploding. Some blackened pieces found along the wreckage trail indicated that whatever caused their blackening was not intense or sustained. A likely explanation is that, late in the break-up sequence, fuel from a ruptured tank ignited.

Weather radar images suggested that the aircraft could have been funnelled between two lines of thunderstorm cells, leaving the crew no alternative but to attempt to fly through a gap between the cells if they wanted to comply with the owner’s visual flight policy. In view of the owner’s attitude to non-visual flight, it seems unlikely that the crew would have deliberately flown into cloud, particularly that associated with thunderstorms.

But the aircraft was flying in close proximity to active frontal thunderstorm cells, and an extreme turbulence encounter was a possibility, even several kilometres from the cloud.

Such turbulence could have thrown the aircraft into an unusual attitude, resulting in excessive airspeed. Its structural limits could then have been exceeded, particularly if there were severe vertical gusts during an attempt to recover.

The practice of the owner and pilotin- command of flying visually whenever possible suggests that the flight would not have continued had they been concerned about the weather ahead. It also adds weight to the possibility that severe turbulence precipitating the break-up occurred in clear air.

Thunderstorm Dangers

This is by no means the first time severe thunderstorm weather has been responsible for breaking up aircraft in flight. Perhaps the best known instance in Australia was the destruction of Ansett-ANA’s Viscount 720, VH-TVC, only minutes after taking off from Sydney in wild weather on a late afternoon flight to Canberra in 1961.

Fourteen months later in the USA, a Northwest Airlines Boeing 720 (the domestic version of the 707) broke up in similar conditions while climbing out of Miami on a flight to Chicago.

In both these instances it seems that the aircrafts’ design strengths were exceeded during attempts to recover at high speed from severe, turbulenceinduced upsets.

But in 1966, a Braniff International Airways BAC-111 (similar in size and configuration to the better known DC‑9), flying from Kansas City to Omaha on a bright moonlit night, was seen by about 300 ground witnesses to break up in clear air as it approached an almost unbroken line of severe thunderstorms.

The line of storms was associated with a powerful cold front, and the crew had elected to remain at about 4000ft AGL while they penetrated what appeared on their radar to be “the lightest area”. Flying normally, still about five miles from the line of cloud, the aircraft was seen to suddenly become a “ball of fire”, before spiralling down into the ground. Two large pieces of the aircraft fell more slowly.

The NTSB investigation found that the BAC-111 encountered an extreme vertical gust which forced it violently upwards, failing its tail assembly. The wing’s ultimate negative design load was then exceeded as the aircraft pitched violently nose-down, rupturing the wing tanks and releasing fuel which immediately ignited. It was apparent that a cold outflow of turbulent air from the advancing storm front had reached out as far as where the break-up occurred.

Some time later the US FAA published an Advisory Circular describing the hazards of thunderstorms and offering guidance to pilots. Among other points it declared: • Potentially hazardous turbulence is present in all thunderstorms, and severe thunderstorms can destroy an aircraft.

• Outside the cloud, shear turbulence has been encountered several thousand feet above, and 20 miles laterally from a severe storm.

In Australia, the Bureau of Meteorology’s Manual of Meteorology, Part 2, points out that frontal and squall line thunderstorms harbour the worst flying conditions, some being potentially destructive to aircraft. Strong vertical gusts (those that present the greatest hazard), can occur anywhere in a storm, but are likely to be most severe near up and down draughts in mature storms.

The danger posed is twofold: Severe loads can be imposed on the aircraft, while violent changes in aircraft attitude can lead to situations in which attempts at recovery exceed the aircraft’s design limits.

While radar can assist in avoiding storm cells, it identifies rain areas only, and severe turbulence can occur well away from the rain echo and the body of the cloud itself.

The overall message for us all is clear – give thunderstorms, particularly those associated with frontal systems and squall lines, a wide berth!