History of Fatigue

The history of fatigue of metals, components, and structures goes back to the 1830s when failures of chains in mines were reported due to dynamic loading, and fatigue testing of these chains was performed for mitigation (Schütz 1996). In association with this, the first wires were invented to avoid the problems with fatigue of the chains. Since then, up until year 2000, around 100,000 papers related to fatigue were published (Schijve 2003, 2009). With so much published literature available, providing a broad and objective historical overview would be highly challenging. Thus, the historical presentation provided here is limited to those aspects that are of most relevance as background for this book. Reference is made to Schütz (1996), Stephens et al. (2001), and Anderson (2005) for a more detailed historical presentation related to fatigue.

The term “fatigue” is apparently first mentioned in the literature in 1854, by an Englishman called Braithwaite. In his paper, Braithwaite describes many service fatigue failures of brewery equipment, water pumps, propeller shafts, crankshafts, railway axles, levers, cranes, and so on. At about the same time many disastrous railroad accidents occurred, such as one on 5 October 1842 when an axle broke at Versailles due to fatigue and the lives of 60 people were lost. Failures of railway axles became a serious problem and as late as in 1887, an English newspaper reported “the most serious railway accident of the week.” In many cases these accidents were due to fatigue failures of axles, couplings, and rails.

In some publications, the fatigue strength in terms of S-N curves is presented as “Wöhler curves” that are named after the work that Wöhler performed in Germany to determine the fatigue strength of railway axles based on fatigue testing in the period from 1860 to 1870. Already in 1858, Wöhler was measuring the service loads on railway axles using self-developed deflection gauges. He also introduced the concept of safety factors, where two sets were needed: one for maximum stress in service in relation to static strength, and the other for allowable stress amplitude under dynamic loading. The safety factors were provided for ensuring design for infinite life. The factors were valid only for un-notched specimens, and fatigue testing was recommended for other geometries. Wöhler presented his test data in tables.