Abstract

Very high cycle fatigue properties of various steels were studied using findings of previous research and laboratory fatigue testing. First, experimental data for more than 550 specimens covering 25 high and medium strength steels were used to investigate the relationships between the applied stress, number of failure cycles, size of defects or inclusions at fracture origins and stress intensity factors. Using the results of the investigation of these data, general conclusions were arrived at for steels as a whole. It was observed that the size of the failure origin can be predicted using strength properties of steels. Existing methods for estimating major parameters such as size of failure origins and stress intensity factors were reviewed, new methods were proposed and their accuracy was verified using experimental data. Also, the possibility of simplifying existing formulae with substitutions for the major parameters was reviewed. Employing these major parameters, new formulae for predicting fatigue strengths of both medium and high strength steels were proposed. Predictions of these proposed formulae were compared with existing well known formulae using experimental data and statistical methods highlighting the simplicity and importance of the proposed formulae. The ability of employing the proposed formulae for predicting, “fatigue strengths that are more close to the real values” as well as “fatigue strengths that are more safe and conservative” was reviewed. Secondly, fatigue properties and failure causes of medium strength – low carbon structural steels that are usually used in civil engineering structures were investigated. For this investigation, 35 smooth specimens of five steels were tested using a rotating bending fatigue tester. It was observed that fatigue failures occur up to around 107 cycles and that the failure originates from the surface. It was found that the formulae proposed are able to predict failures of these medium strengths steels. Slopes of stress life curves in the very high cycle fatigue regions were well predicted by these proposed formulae while the predictions were fairly aligned with values suggested in previous research. Finally, recommendations were given for employing suitable prediction methods considering safety and importance of components and structures.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call