Steel construction - Fatigue verification according to Eurocodes

A structure is made up of constructive details: it can fail when at least one of these constructive details reaches one of the ultimate limit states. Among these limit states is the fatigue limit state. This state reflects progressive cracking due to repeated loading over a very large number of cycles over time. Once a certain level of cracking has been reached, failure may occur.

Crack initiation necessarily starts from a defect. In the case of a part with a smooth appearance, the crack may originate from microscopic defects at the grain scale, or from edge defects due to flame cutting, drilling, etc., or it may originate from the surface of the part.

In the case of parts with welds, it's more likely to be a defect in the weld bead that initiates a crack (solid or gaseous inclusion, retassure, crack, etc.). These defects, combined with the notches caused by the more or less abrupt change in geometry due to the weld bead, are more significant than those in parts without welds.

Fatigue verification, which serves to protect against cracking, can prove decisive in the case of bridges, runways, very slender structures exposed to wind, or hydraulic structures. It is unnecessary for ordinary buildings subject to quasi-static actions such as wind or snow.

For fatigue testing, reference is generally made to the standard NF EN 1993-1-9 Design of steel structures – Part 1-9: Fatigue". Other Eurocodes, however, provide additional details of this verification for structures that may be susceptible to fatigue, such as bridges, towers, masts, chimneys or runways.

The calculation of crack propagation is discussed, but without much detail, in the NF EN 1993-1-10 or rather in its reference document .

Having to refer to different parts of Eurocode 1 for fatigue loads, and to different parts of Eurocode 3 for the verification...