Fatigue in Mechanical Components: Understanding SN Curves, Endurance Limit, and Life Prediction

Fatigue and Life Prediction in Brief

Fatigue failure happens when components experience time-varying loading, causing crack formation and growth that can lead to fracture at stresses well below a material’s ultimate strength. This video introduces fatigue as a three-stage process: crack formation, crack growth, and fracture. It explains how fatigue tests generate SN curves by plotting cycles to failure against the stress range on a log scale, enabling life predictions for a given stress range. The discussion covers the endurance limit, high cycle versus low cycle fatigue, and the influence of mean stress using the Goodman diagram. It also highlights variability in fatigue data, the need to shift published SN curves to account for scatter, and practical tools like rainflow counting and Miner’s rule for cumulative damage. The talk closes with alternate approaches when cracks are present and a nod to real-world complexities.

Fatigue and Life Prediction in Engineering

Fatigue failure arises when loading varies with time, causing cracks to initiate at surfaces and stress concentrations and to propagate until fracture. The video presents fatigue as a three‑stage process: crack formation, crack growth, and final fracture. To assess fatigue risk, engineers perform constant amplitude stress cycle tests on test pieces to determine cycles to failure, N, for various applied stress ranges, S. By plotting N versus S on a log horizontal axis, an SN curve is obtained which can be used to predict the number of cycles to failure for a given stress range. In practice, SN curves for many materials are published in engineering codes, reducing the need for new testing. A key feature is the endurance limit, a horizontal tail in the SN curve at a very large number of cycles, where fatigue life is theoretically infinite if the stress range stays below this value.

High cycle fatigue vs low cycle fatigue is another important distinction. High cycle fatigue involves low stresses and elastic deformation, with life extending to more than 10 000 cycles. Low cycle fatigue involves higher stresses that exceed yield, causing both elastic and plastic deformation and typically requiring strain-based design tools such as Coffin-Manson relations rather than SN curves.

Mean stress effects and the Goodman diagram illustrate how nonzero mean stresses can shorten fatigue life. The Goodman diagram relates mean stress and stress amplitude to determine whether a loading condition could produce infinite life. While it is a useful note for infinite life assessment, it does not directly yield a fatigue life estimate. For complex loading spectra, techniques like rainflow counting reduce the spectrum to equivalent simple cycles, and Miner’s rule sums the damage contributions from each cycle, D = sum(n_i / N_i). If D reaches 1, fatigue failure is predicted. The video also mentions shifting SN curves downward to account for data variability, typically by several standard deviations, to reduce the chance of overly optimistic life predictions.

Real-world complexities and alternatives include nonzero crack presence, where linear elastic fracture mechanics provides a way to estimate critical crack sizes and growth to fracture. Overall, the video outlines a practical framework for fatigue life prediction, balancing conservative design and realistic material behavior while acknowledging variability and mean-stress effects.