Structural Fatigue and Aircraft Safety

In 1941 the Beaufort Division of the Aircraft Production Commission requested advice from the Aeronautical Research Laboratories on the use of local alloy steels for welded tubular components such as engine mountings, where fluctuating loads could cause possible fatigue effects. Its investigation led ARL to think more generally about fatigue in the aircraft structure itself. The increasing size of aircraft structures was leading to very low natural frequencies of oscillation of the structures, and this, together with the higher speeds, was suspected to be leading to coupling with turbulence and other loads which could begin to cause fatigue effects.

In 1945 a Stinson airliner (modified) crashed in Australia, killing 10 people, and it was proved beyond doubt that a fatigue failure had occurred in a main welded tubular spar of the wing. ARL staff, and engineers of the Department of Civil Aviation (DCA), collaborated to assess the general significance of the fatigue question. When an extensive search of overseas literature revealed no sound basis for control of the hazard, a broad program of investigation was put in train.

The investigations included the statistical measurement of flight loads, the fatigue resistance of typical aircraft structures, basic studies of crack initiation and propagation, the dynamics of aircraft as elastic structures under load, and the formulation of methods for estimating the statistical 'safe life'.

In 1947 ARL scientists reported progress in a paper by H. A. Wills to the I. E. Aust,[53] and following further intensive research, presented a comprehensive review to the Second International Aeronautical Conference (New York 1949).[54] It discussed factors contributing to increased fatigue hazards, outlining by numerical example a procedure for estimating flying hours to failure (given adequate fatigue test data for the aircraft concerned).

The crash of a Martin 202 in the United States with a self evident fatigue cause after only 1400 hours flying time, gave strength to the ARL views and led to their adoption by the international community.

Because of its early entry into the field, through the Stinson failure, and the systematic collection of data, ARL,[55] established a lead in the work and was thus to become, and remain, an international leading centre for this kind of research. The work was supported by the civil aviation authorities because at that time the pattern of usage of Australian aircraft was such as would generate a higher risk of fatigue failure than in other countries.

An intensive and definitive programme of research was conducted on full scale structures. Preliminary work was done on the wings of the CA 12 Boomerang, but the major work was a massive testing program and analysis of 105 wings of Mustang aircraft. The sound statistical base thus generated permitted the conjunction of structural theory, stress analysis, and metallurgy to be brought together in a practical way for the prediction of the safe life of aircraft, in service inspection and the derivation of maintenance schedules.

Australian Academy of Technological Sciences and Engineering