Fatigue Crack Growth and Paris Law Explained

Fatigue crack growth analysis predicts how a pre-existing crack advances cycle by cycle under repeated loading, complementing S-N curve fatigue analysis by explicitly answering how long a component can safely operate once a flaw is present. This guide covers the stress intensity factor range, the three regions of the fatigue crack growth curve, the Paris Law equation and its integration into fatigue life, load ratio effects, and how these tools underpin damage-tolerant design and inspection planning.

From S-N Curves to Damage-Tolerant Design

Classical S-N curve fatigue analysis treats life as the total number of cycles to failure for a nominally flaw-free specimen, and is dominated by the crack initiation stage — the accumulation of microstructural damage that eventually nucleates a fatigue crack. This works well for safe-life design philosophy, where components are simply replaced before their statistically expected life is reached. It says little, however, about how a component with a known or postulated existing flaw — a weld defect, an inclusion, a corrosion pit, or an in-service crack found during inspection — will behave going forward. Fatigue crack growth analysis, built on fracture mechanics rather than S-N statistics, fills that gap directly: given a starting crack size, it predicts how many cycles remain before the crack reaches a critical, unstable size, forming the basis of damage-tolerant design used throughout aerospace, pressure equipment, and structural steel codes.

The Stress Intensity Factor Range, ΔK

Fatigue crack growth correlates strongly with the cyclic range of the crack-tip stress intensity factor, not the applied stress range alone, because ΔK captures the combined effect of stress, crack size, and geometry in a single parameter, exactly as the static stress intensity factor K does in monotonic fracture mechanics.

ΔK = Kmax - Kmin = Y Δσ √(πa)

where:
  Δσ = σmax - σmin   (applied stress range)
  a    = current crack length
  Y    = dimensionless geometry factor (crack/component geometry)

Two components with different absolute stress and crack size but
the same ΔK will grow their cracks at the same rate.

The Three Regions of Fatigue Crack Growth

Plotting crack growth rate, da/dN, against ΔK on logarithmic axes produces a characteristic curve with three distinct regions.

Region I: Near-Threshold

At low ΔK, crack growth rate falls very steeply and approaches a practical lower asymptote, the fatigue crack growth threshold, ΔK_th, below which cracks grow so slowly they are generally considered non-propagating for engineering purposes. This threshold is sensitive to load ratio R, microstructure, and environment, and represents an important design limit: a component can in principle tolerate an existing crack indefinitely provided the applied ΔK stays below ΔK_th.

Region II: The Paris Regime

Across the mid-range of ΔK, the log-log curve is essentially a straight line, described by the Paris Law. This is the largest region of the curve and, for most structural components, accounts for the great majority of the fatigue crack propagation life, which is why Region II receives the most attention in practical fatigue crack growth assessments.

Region III: Unstable Growth

As K_max approaches the material’s fracture toughness (Kc or K1c), growth rate accelerates sharply above the Paris Law trend line, culminating in final unstable fracture. This region is typically of interest only when very few cycles remain, since crack growth here consumes a small fraction of total propagation life, but it defines the upper bound — the critical crack size — used to close out a fatigue crack growth calculation.

The Paris Law Equation

da/dN = C (ΔK)m

where:
  da/dN = crack growth rate per cycle (e.g. mm/cycle or in/cycle)
  ΔK   = stress intensity factor range (e.g. MPa√m or ksi√in)
  C, m  = material constants, determined experimentally
          per ASTM E647

Typical range: m ≈ 2-4 for most structural metals
              (reported range across material classes: 2-7)

C and m are empirical constants specific to a given material, environment, temperature, and load ratio, and are found from crack growth testing per ASTM E647 or from published material property databases. Because C’s numerical value depends on the units used and on the value of m, C should never be quoted without its accompanying units and the m value it was fitted alongside.

Integrating Paris Law for Fatigue Life

Design and inspection planning require total cycles, not just instantaneous growth rate, so the Paris Law is integrated from an initial crack size, a₀, to a final or critical crack size, a_f.

For m ≠ 2:

N = [ af(1 - m/2) - a0(1 - m/2) ]
    -----------------------------------------------------
    C (Y Δσ)m π(m/2) (1 - m/2)

For m = 2 (special case, logarithmic form):

N = ln(af / a0)  /  [ C Y² Δσ² π ]

where:
  a0 = initial crack size (from inspection sensitivity
         or a postulated manufacturing flaw)
  af = final/critical crack size (from K1c or Kc,
         ac = (1/π)(K1c / (YΔσ))²)

In practice, Y often varies with crack length for real component geometries, and the integral must be evaluated numerically rather than in closed form; nonetheless, the closed-form result above illustrates the strong sensitivity of predicted life to the assumed initial crack size a₀ and to the exponent m — small changes in a₀ or in the assumed material constants can shift calculated life by a wide margin, which is why probabilistic treatments are increasingly used alongside a single deterministic estimate.

Load Ratio Effects: Walker and Forman Equations

The load ratio, R = σ_min/σ_max = K_min/K_max, has a strong effect on crack growth rate that the basic Paris Law does not capture, since Paris Law depends only on the range ΔK, not on the mean stress level. Higher R generally increases growth rate and lowers the effective threshold. Two widely used extensions address this:

da/dN = C [ ΔK / (1 - R)(1-γ) ]m

where γ is a material-specific fitting constant (γ = 1
recovers the basic Paris Law with no R-dependence)

da/dN = C (ΔK)m / [ (1 - R) Kc - ΔK ]

Growth rate rises without bound as ΔK approaches
(1 - R)Kc, modelling the Region III upturn directly.

Practical Application: Damage-Tolerant Design

Damage-tolerant design accepts that flaws may exist or develop in service and designs the inspection and maintenance programme around fracture mechanics predictions rather than assuming a flaw-free structure. The general workflow:

• Establish an initial crack size a₀, typically the largest flaw a chosen non-destructive inspection method could plausibly miss (its probability-of-detection limit), not simply the smallest flaw it could theoretically find.
• Determine the critical crack size a_f from the material’s fracture toughness and the applied stress at the location of concern.
• Integrate the Paris Law (or an R-corrected variant) to calculate cycles from a₀ to a_f.
• Set the inspection interval at a fraction of this calculated life, commonly with a safety factor of 2 or more on life, or an equivalent factor on crack size, to account for uncertainty in C, m, Y, and initial flaw size.
• Re-inspect and, if a crack is found, re-evaluate remaining life from the newly measured crack size rather than the original assumption.

Limitations of Paris Law

Frequently Asked Questions

Recommended Reference Books

Disclosure: MetallurgyZone participates in the Amazon Associates programme. If you purchase through these links, we may earn a small commission at no extra cost to you. This helps support free technical content on this site.

Further Reading