Frank Yeend was an officer of the Air safety Investgation Branch of the DCA from 1953 until 1975, and Head of the Branch during the last three years of this term. During these 22 years he was involved in many field investigation of aircraft accidents, and in the Boards of Accident Inquiry subsequently established.
This is Frank's recollections of the investigation of a major accident involving the Vickers Viscount VH-RMI.

 

 

In the early afternoon of September 22, 1966 I was in our Melbourne office and received a telephone call from the Regional Director in Queensland, Mike Seymour, that an Ansett-AN.A Viscount aircraft had crashed near Winton in Queensland and it was believed there were no survivors. My then chief, David Graham, was interstate on duty at the time and so I set about organising the nucleus of an investigation team. We set off in the late afternoon, first of all to Brisbane to collect additional members and equipment, and then to Winton, arriving at 3.45 a.m. on the following morning.
Winton is predominantly a sheep farmers' town at about dead centre in the State of Queensland, with a population of a little over 1,000 people. It lays claim to being the home of "Waltzing Matilda" as Banjo Patterson wrote the song on a nearby station and it had its first airing in the town. It also had quite a good aerodrome, comprising two airstrips on flat open terrain.

VH-RMI had departed from Mount Isa just after mid-day for a scheduled flight to Brisbane, with a landing at Longreach en route. The weather was fine with broken cloud and the captain elected to cruise at 16,000 feet. The flight proceeded without alarm for some 45 minutes and then the aircraft called Longreach to say it was on an emergency descent. In subsequent communications the crew reported fire warnings in two engines and a little later that there was a visible fire in No. 2 engine and that they were diverting to land at Winton. Nothing further was heard until people reported black smoke at a point 13 1/2 miles west of the Winton aerodrome.

The investigation party was taken to the accident site at first light and we were presented with a scene of absolute devastation. The wreckage of the aircraft, in many thousands of pieces, was strewn through light gidgea scrub over a distance of 1.2 miles and the whole area had been ravaged by fire. The wreckage recovery and examination task was obviously going to demand a herculean effort, but it was done tremendously well over many months by the Group leaders Don Whalley, Colin Torkington, Ray Broughton and Jim Doubleday.

As the investigation progressed, we also brought in assistance from Ansett-A.N.A, the British Aircraft Corporation, Rolls Royce and Dowty Rotol. The Operations Group was headed up by Colin Beech, with assistance at times from people like lan Leslie, Bob Whitecross, Russ Watts and Paul Choquenot. David Graham also spent some time with us in an oversight role or assistance to me as the Investigator-in-Charge. We had a little trouble early on with the BAC senior representative, who decided to ignore the rules on security of information, but a quiet word in the right place by David had him recalled and replaced.

The work of the Operations Group went ahead in the usual manner, interviewing eye or hearing witnesses, collecting and analysing the relevant operational documents, compiling the record of radio communication, assessing the weather conditions, looking at the flight crew histories and reviewing the aircraft load and fuel supply. Beyond that, this Group could only wait upon the development of the engineering and medical evidence before setting about a reconstruction of the flight from Mount Isa to the accident point.

 

The wreckage examination at the scene had to be undertaken with meticulous care, first of all, to reconstruct the sequence of what was obviously an in-flight failure. The port outer wing and No. I engine, although close to the main impact area, were located in a distinctly separate area of impact. We already knew of the pilot's report of an in-flight fire and so a difficult, but very important aspect of the examination was to distinguish between the evidence of in-flight fire and the effects of the post impact fire. Although the accident site was in an isolated area and so we could work largely untroubled by public visitors, working in the open for long hours, even at this time of the year, was very trying.

After two weeks of this we decided to pick up all the vital parts and move them by truck to Melbourne. Here we had rented an old wool store in Footscray and the task of putting the pieces back in their conventional relationship began in earnest. It was not long before the hard evidence of an in-flight fire commencing at the rear of the No. 2 engine, progressing into the wheel bay and then into the adjacent fuel cell, became very apparent. At the same time, and with the help of the Aeronautical Research Laboratories, the port wing main spar became the focus of close study.

It emerged that the upper boom of this single spar wing had failed when its strength had been reduced by heat to only about 12% of the normal strength. The lower spar boom then failed under overload and the wing separated when the aircraft was at a height of about 4,000 feet. The real tragedy, of course, was that the aircraft was then within 5 minutes of landing at Winton. It may have been possible to put the aircraft down on the open country near the accident site, but there was no way the pilots could have known how extensive was the fire in the aircraft and thus how immediate was the emergency.

 

The investigation of this accident took a little more than six months and then there were the lengthy processes of the Board of Accident Inquiry under the chairmanship of Sir John Spicer, assisted by Bert Ritchie of Qantas, Frank Ball of TAA, Col Griffin of Ansett-A.N.A. and Tom Air of the Department of Supply. The Board first sat on April 2l, 1967, some 7 months after the accident and the counsel assisting were Charles Sheahan QC and Alan McCracken, both of Brisbane. Wally Campbell QC - later Governor of Queensland - appeared for Ansett-A.N.A., Gordon Samuels - later Governor of N.S.W. - for the British Aircraft Corporation, Ned Williams QC for the Department of Civil Aviation, Bill Crockett QC for Godfrey Engineering Products and Arthur Pearce for the Australian Federation of Air Pilots.

For the most part the inquiry was conducted in Brisbane, with short adjournments to Melbourne and Winton, to view the wreckage, the accident scene and to hear evidence from some of the local witnesses. This was the first accident investigation in Australia that had been aided by information from a flight data recorder. It was a rather crude seven parameter scratch recorder and had been severely damaged by fire and impact but, nevertheless, we obtained sufficient information to be able to accurately reconstruct the final flight path of the aircraft right up to the accident point.

In this Inquiry a recommendation of mine, that the investigation team should support the counsel assisting the Board rather than the counsel for the Department of Civil Aviation, was adopted. In earlier inquiries there had been some conflicts of interest and some compromise of our position as independent investigators. The new arrangement worked well but, again, I had to spend several days in the witness box, outlining the scope of our investigation and explaining the logic of our findings.

Having established that the fire in the air associated with No. 2 engine caused the wing failure and the subsequent impact with the ground, of course, we had to discover the source of the fire. Because the Viscount had a pressurised and air-conditioned cockpit and cabin, there was a need for a continuous supply of air under pressure to the under-floor conditioning units. The pressure air was supplied by air pumps or cabin blowers, taking power from 3 of the 4 engines, including No. 2 engine. These air pumps operate at very high r.p.m., with very close clearances and so the bearings are continuously fed with an oil supply via an oil-metering unit, bolted to one end of each pump.

The wreckage examination revealed that the fire in flight had started in the cabin blower powered by No. 2 engine. Apparently a minor metal failure in one of the pump lobes had set up a vibration which caused the 5 studs and nuts securing the oil meter to back-off, allowing one end-bearing of the lobes to come free. This induced very high friction heat in the blower which ignited the oil being supplied to the bearing. The fire then spread into the adjacent wheel bay and from there into the front cell of the fuel tank between Nos. 1 and 2 engines. This fuel cell was more than half full of fuel and it lay immediately adjacent to the wing main spar.

The initial fault in the cabin blower and the progression of the fire most probably commenced not long after the take-off at Mount Isa, but the cabin blowers are installed rear of the fireproof bulkhead and thus outside the fire sensing zone of the engine compartment. The failure of one cabin blower of the three operating would not have affected the air supply significantly and the engine itself would continue to operate normally, at least until the fire had grown to large proportions. In the latter stages of the flight, the fire and smoke on the port side of the aircraft became visible from the cabin, but by this time the softening of the wing spar would have progressed to the point where a catastrophic failure was imminent. The attachment of the No. 2 blower oil metering unit was effected by 5 threaded studs in the body of the unit upon which five castellated nuts held secure a cover plate.

 

In turn, these castellated nuts are secured by a copper locking, wire passing through each to prevent unwanted rotation of the nuts or the studs. In the wreckage examination only one stud remained and none of the nuts, but obviously there was a strong presumption that, although there was a severe vibration set up by the blower lobe failure, the loss of the studs and nuts probably occurred because the required locking wire had not been put in place during the last overhaul of the blower.

The Board of Inquiry listened patiently to all of this rather complex engineering evidence, but eventually submitted a report to the Minister and through him to Parliament, which adopted our conclusions entirely. This report was tabled in Parliament on October 17, 1967. In typical fashion, the Chairman lavished great praise on the investigating officers and then turned round to critisize our presentation of evidence on a matter which was not only minor, but based upon a complete misconception of the situation. The Minister for Civil Aviation at the time was Reg Swartz and so we had him point out this misconception in his tabling statement.

As a result of this accident several changes to operating and maintenance procedures were made by Ansett-A.N.A and by the contracting organisation responsible for the periodic overhaul of the cabin blowers. The accident also emphasised the need for airline aircraft to carry cockpit voice recorders, as well as flight data recorders, particularly to cover accidents where the flight crew are not survivors. Within a couple of years most airline aircraft in Australia were fitted with this type of equipment.