Menu
Metal Fatigue: S-N Curves, Crack Initiation and Prevention

Metal Fatigue: S-N Curves, Crack Initiation and Prevention

How metal fatigue works: crack initiation, S-N curves, the endurance limit, Goodman mean-stress correction, and how to prevent and detect fatigue cracks.
Metal Fatigue: S-N Curves, Crack Initiation and Prevention

Metal Fatigue: S-N Curves, Crack Initiation and Prevention is the study of progressive, localised cracking in a component subjected to fluctuating stress at levels well below its ultimate tensile strength. A shaft, weld or bracket that would carry a static load indefinitely can still fail after enough load reversals, because each cycle advances a microscopic crack a little further. Fatigue is the dominant failure mode for rotating and reciprocating machinery, and understanding it is the difference between a planned overhaul and an unplanned catastrophic fracture.

What metal fatigue is

Fatigue damage accumulates under cyclic loading even when the peak stress stays comfortably inside the elastic range. The key point is that the stress that causes failure over millions of cycles can be a fraction of the yield strength. Because the damage is localised at the surface and at geometric features, a part can look perfectly sound right up until it breaks. Load sources include rotating bending, torsional reversals, pressure pulsations, thermal cycling and vibration.

The three stages of fatigue failure

A fatigue fracture almost always develops in three distinct stages:

  • Crack initiation. A microcrack nucleates at a stress concentration, usually a surface feature such as a fillet, keyway, weld toe, tool mark, inclusion or corrosion pit.
  • Crack propagation. The crack grows a small increment on each cycle. On the fracture face this leaves beach marks (macroscopic bands visible to the eye, reflecting changes in load) and striations (microscopic ridges, one roughly per cycle, seen under an electron microscope).
  • Final fracture. When the remaining cross-section can no longer carry the peak load, the part fails suddenly in a single overload event, often leaving a rough, fibrous or crystalline final-fracture zone.

Reading these features on a broken part tells you the origin, the direction of growth and whether the load was high or low relative to the section.

The S-N (Wohler) curve and the endurance limit

The classic fatigue characterisation is the S-N curve, first mapped by August Wohler. It plots stress amplitude (S) against the number of cycles to failure (N) on a logarithmic scale. Higher stress means fewer cycles to failure. For many ferrous alloys the curve flattens into a horizontal asymptote: below a certain stress, the endurance limit (or fatigue limit), the part survives an effectively infinite number of cycles. Many non-ferrous alloys, notably aluminium, show no such plateau; their curve keeps sloping downward, so engineers quote a fatigue strength at a defined cycle count instead.

Material classTrue endurance limit?Approx. fatigue ratio (Se/UTS)Reference life
Wrought steelsYes~0.5 (capped near 700 MPa)10^6 to 10^7 cycles
Cast ironYes~0.410^6 to 10^7 cycles
Titanium alloysYes~0.45 to 0.610^7 cycles
Aluminium alloysNo~0.3 to 0.4quoted at 5x10^8 cycles
Copper alloysNo~0.3quoted at 10^8 cycles

The ratios above are engineering approximations for polished laboratory specimens; real components need correction factors.

Stress concentration, surface finish and mean stress

Three effects push a real part below its textbook fatigue strength:

  • Stress concentration. A notch, hole or sharp corner multiplies local stress by a factor Kt. Fatigue cracks nucleate exactly where Kt peaks.
  • Surface finish. Rough machining, forging skin and scale all lower fatigue strength because initiation is a surface phenomenon. A ground or polished finish raises it.
  • Mean stress. A tensile mean stress reduces the allowable alternating stress. The modified Goodman relation captures this: sigma_a / Se + sigma_m / Su = 1, where sigma_a is the alternating amplitude, sigma_m the mean stress, Se the endurance limit and Su the ultimate strength. Soderberg (using yield strength) is more conservative; Gerber uses a parabola.

High-cycle versus low-cycle fatigue

Fatigue splits into two regimes. High-cycle fatigue (roughly above 10^3 to 10^4 cycles) is stress-controlled, the material stays largely elastic, and the S-N curve governs design. Low-cycle fatigue (below that range) involves gross plastic strain each cycle, as in thermal transients or occasional heavy overloads; here the strain-life (Coffin-Manson) approach is used instead of stress-life. Knowing which regime a component lives in decides whether you design against stress amplitude or plastic strain range.

Preventing fatigue failures

Because initiation dominates fatigue life, prevention concentrates on the surface and on geometry:

  • Eliminate stress raisers: blend welds, deburr edges, avoid sharp re-entrant corners and abrupt section changes.
  • Use generous fillet radii at shoulders and transitions to spread stress.
  • Apply shot peening or roller burnishing to induce compressive residual stress at the surface, which suppresses crack initiation.
  • Specify a good surface finish and control machining marks in highly stressed zones.
  • Protect against corrosion; a corrosive environment removes the endurance limit entirely and can trigger stress corrosion cracking and corrosion fatigue.

Detecting fatigue cracks in service

Surface-breaking fatigue cracks are found with non-destructive testing. Liquid penetrant testing reveals cracks on any non-porous material, while magnetic particle testing is faster on ferromagnetic parts and picks up slightly sub-surface indications. On a rotating asset, vibration trending often flags a growing crack before it breaks the surface. Feeding NDT results, inspection intervals and crack-growth history into a CMMS keeps the evidence in one place; teams using Fabrico schedule recurring inspections and attach findings to the asset record so a rising trend triggers action rather than a surprise failure. Book a Fabrico demo to see that inspection workflow.

Frequently Asked Questions

Why do parts fail below their tensile strength?

Because fatigue is cumulative. Each load cycle advances a microscopic crack a tiny amount, so given enough cycles a stress well under the static strength can grow a crack to critical size and cause sudden fracture.

Do all metals have an endurance limit?

No. Most steels, cast irons and titanium alloys show a true endurance limit below which life is effectively infinite. Aluminium and many other non-ferrous alloys do not, so a fatigue strength is quoted at a stated cycle count instead.

What are beach marks and striations?

Both are marks on the fracture surface left by crack propagation. Beach marks are macroscopic bands visible to the eye that reflect changes in loading. Striations are microscopic ridges, roughly one per cycle, seen only under an electron microscope.

How does shot peening improve fatigue life?

It plastically deforms the surface layer, leaving it in residual compression. Since fatigue cracks start under tension at the surface, that compressive layer must be overcome first, delaying initiation and raising fatigue strength.

Latest from our blog

Define Your Reliability Roadmap
Validate Your Potential ROI: Book a Live Demo
Define Your Reliability Roadmap
By clicking the Accept button, you are giving your consent to the use of cookies when accessing this website and utilizing our services. To learn more about how cookies are used and managed, please refer to our Privacy Policy and Cookies Declaration