A Guide to the S-N Curve — Stress vs. Cycles-to-Failure, Fatigue Limits, and What 'Fatigue-Resistant' Actually Means
A guide explaining the S-N curve — the fundamental engineering plot of applied cyclic stress against the number of cycles a component survives before fatigue failure — and how fatigue life engineering, fatigue limits, and stress concentration underlie every claim about a component's resistance to cyclic loading. Shivam Forge, Rajkot, India. Call +91-9265772827.
A component can fail under cyclic loading at a stress level far below what it would ever fail at under a single, static application of that same load — a phenomenon so counterintuitive when first encountered, and so consequential in practice, that it drove the development of an entire branch of mechanical engineering. Fatigue failure occurs through a progressive process: a microscopic crack initiates, typically at a surface stress concentration (a machining mark, a material inclusion, a fillet radius, a forging lap), and grows incrementally with each load cycle until the remaining cross-section can no longer carry the load and fractures suddenly, often with little visible warning beforehand. The S-N curve (S for stress, N for number of cycles) is the foundational tool for characterizing this behaviour: test specimens of a given material and condition are subjected to cyclic loading at a series of different stress amplitudes, the number of cycles each specimen survives before failure is recorded, and the results are plotted with stress amplitude on one axis and cycles-to-failure (typically on a logarithmic scale, since fatigue life spans orders of magnitude) on the other. The resulting curve slopes downward — higher applied stress produces shorter fatigue life — and for many steels, the curve eventually flattens into a horizontal asymptote called the fatigue limit or endurance limit, a stress level below which the material can theoretically sustain cyclic loading indefinitely without fatigue failure. This single curve underlies every claim made anywhere about a component being 'fatigue-resistant': it is only meaningful relative to a specific applied stress amplitude and required design life, since a material's fatigue performance is not a single number but an entire curve relating stress to survivable cycle count.
Stress amplitude (or stress range) is plotted on the vertical axis, and cycles-to-failure on a logarithmic horizontal axis, since fatigue life at different stress levels for the same material can span from a few thousand cycles to tens of millions — a linear scale would compress the low-cycle region into illegibility.
Many carbon and low-alloy steels exhibit a true fatigue limit — the S-N curve flattens to a horizontal asymptote, typically established by around 10^6–10^7 cycles, below which the material theoretically survives indefinitely. Many non-ferrous alloys (notably aluminium) do not exhibit a true flat limit and instead show continuously declining strength with cycle count, requiring a specified design life rather than an assumed infinite life.
High-cycle fatigue (typically beyond roughly 10^4–10^5 cycles) involves stress levels below yield strength and elastic deformation dominating each cycle; low-cycle fatigue involves higher stress with significant plastic deformation each cycle and correspondingly much shorter life, governed by different analytical approaches (strain-life rather than stress-life methods).
Fatigue life shows genuine scatter even among nominally identical specimens tested at the same stress level, since crack initiation is sensitive to microscopic material and surface variation; published S-N curves and design allowables typically represent a statistically defined lower bound (such as a 95% survival probability) rather than a mean value, appropriately biasing design toward safety.
A component's actual fatigue performance is governed less by the base material's S-N curve alone than by the local stress concentration at fillets, holes, thread roots, and surface features, which amplifies nominal stress well above the applied load — designing generous fillet radii and avoiding sharp geometric transitions is often more consequential to fatigue life than material grade selection alone.
Since fatigue cracks overwhelmingly initiate at the surface, surface condition materially affects real-world fatigue performance relative to a polished laboratory test specimen — rough machining marks reduce fatigue life, while compressive residual stress from shot peening or fillet rolling measurably improves it by delaying crack initiation.
Forging's continuous, uninterrupted grain flow — grain structure that follows a component's contour rather than being cut across by machining from bar or plate stock — reduces the internal discontinuities that can act as fatigue crack initiation sites, a genuine mechanical contributor to why forged components are widely specified for cyclically loaded applications over cast or machined-from-billet alternatives.
A meaningful fatigue design conversation starts by defining the actual expected stress amplitude and required number of cycles the component will see in service, then checking that combination against the material's S-N curve with appropriate safety factor — a component correctly described as fatigue-resistant for one loading and cycle-count combination can fail well short of expectations under a different one.
Fatigue failure is, in a real sense, the opposite of what intuition suggests strength should mean: a component can survive a single application of a given load comfortably, well within its static yield and tensile strength, and still fail catastrophically after enough repetitions of a much lower load applied and removed over and over again. This behaviour — first systematically studied in the mid-19th century after a series of railway axle failures that static strength calculations of the era could not explain — is why fatigue engineering exists as a distinct discipline from static strength design, and why the S-N curve, the plot relating applied cyclic stress amplitude to the number of cycles a material survives before fracture, is the foundational tool underlying it.
Constructing an S-N curve requires testing a series of nominally identical specimens, each cycled at a controlled, constant stress amplitude until fracture, with the cycle count at failure recorded for each specimen and each stress level. Plotted with stress amplitude on one axis and cycles-to-failure on a logarithmic axis (necessary because fatigue life at different stress levels for the same material can span from a few thousand cycles to tens of millions), the resulting curve slopes downward from left to right — unsurprisingly, higher applied stress produces shorter survivable life. What makes the curve particularly useful for many carbon and low-alloy steels is that it eventually flattens into a horizontal asymptote, commonly established by testing out to roughly 10^6 or 10^7 cycles, called the fatigue limit or endurance limit: a stress amplitude below which the material can theoretically withstand indefinite cyclic loading without fatigue failure. Not every material behaves this way — aluminium alloys and several other non-ferrous metals typically show continuously declining fatigue strength with increasing cycle count rather than a true flat asymptote, meaning fatigue design for these materials must specify a target design life rather than relying on an assumed infinite-life stress threshold.
The practical value of the S-N curve concept extends well beyond the laboratory test that generates it, because it reframes what 'fatigue-resistant' actually means for a real component: fatigue performance is never a single number, it is an entire relationship between applied stress and survivable cycle count, and a component is only meaningfully fatigue-resistant relative to its own actual expected loading and required service life. A material with excellent fatigue properties in laboratory testing on a smooth, polished specimen can still fail prematurely in service if the actual component geometry introduces stress concentrations — fillets, holes, thread roots, abrupt section changes — that locally amplify nominal stress well above what the bulk S-N curve was characterized against, or if surface condition (machining marks, corrosion, decarburization) provides ready crack initiation sites the laboratory specimen didn't have. This is precisely why forged components' continuous, contour-following grain flow, generous fillet radii in well-designed forging geometry, and surface treatments like shot peening (which introduces beneficial compressive residual stress at the surface, delaying crack initiation) are genuine, mechanically grounded contributors to real-world fatigue performance — they address the actual crack-initiation physics the S-N curve is fundamentally describing, not just a marketing claim layered on top of it.
Understanding the S-N curve is the necessary foundation for any genuine engineering conversation about fatigue life, cyclic loading capacity, or fatigue-resistant material and design claims — a conversation that should always start from the component's actual expected stress amplitude and required cycle count, checked against the specific material's characterized S-N behaviour with an appropriate safety margin. Shivam Forge's engineering team supports this conversation directly: material grade selection, forging geometry review for stress concentration, and surface treatment recommendations (including shot peening) matched to your component's actual cyclic loading profile. Contact our engineering team at +91-9265772827 or sales@shivamforge.com with your drawing and loading requirement for a manufacturability and fatigue design review.
An S-N curve is a plot, developed from fatigue testing, showing how many cycles of a given applied stress amplitude a material can survive before fatigue failure — higher applied stress produces shorter survivable life, and the curve is the fundamental engineering tool for predicting fatigue performance rather than relying on static strength values alone.
No. Many carbon and low-alloy steels exhibit a true fatigue (endurance) limit — a stress level below which the S-N curve flattens and the material can theoretically endure indefinite cycling. Many other materials, notably aluminium alloys, show continuously declining fatigue strength with increasing cycle count and never fully flatten, meaning a specific design life must be defined rather than assuming indefinite fatigue resistance.
Applied stress is typically highest at the surface for bending and torsional loading, and surface features — machining marks, corrosion pitting, material inclusions breaking the surface, geometric stress concentrations like fillets and holes — provide the microscopic stress risers where a fatigue crack most readily initiates. This is why surface finish and residual stress condition materially affect real-world fatigue life relative to bulk material properties alone.
A geometric feature like a fillet, hole, keyway, or thread root locally amplifies nominal applied stress by a factor (the stress concentration factor) that can be several times the nominal value, meaning fatigue cracking initiates at these features at applied loads well below what the S-N curve would predict for uniformly stressed material — generous fillet radii and smooth geometric transitions are often the single most effective fatigue-life design lever available.
Yes. Our engineering team can review component geometry for stress concentration and fatigue-relevant design features, recommend material grade and surface treatment (such as shot peening) matched to your application's expected stress amplitude and cycle count, and provide mechanical property documentation supporting your own fatigue analysis. Contact us with your drawing and loading requirement to discuss.
Why Choose Shivam Forge
Shivam Forge delivers precision hot-forged components from our integrated Shapar, Rajkot facility — covering forging, CNC machining, heat treatment, and quality inspection under one roof.
