Basquin's Law¶
An empirical power-law segment relating cyclic stress to fatigue life under specified material, loading, and failure conditions.
Core Idea¶
Basquin's law describes an empirical segment of the relation between repeated stress and fatigue life by a power law. In a common modern fully reversed stress-amplitude convention,
where \(\sigma_a\) is stress amplitude, \(N_f\) is cycles to a defined fatigue-failure endpoint, $2N_f$ is reversals, \(\sigma_f'\) is a fitted fatigue-strength coefficient, and \(b\) is the fitted slope on log stress versus log reversals. Equivalently, \(N_f=\tfrac12(\sigma_a/\sigma_f')^{1/b}\). Since \(b\) is negative, higher stress means fewer predicted cycles within the calibrated regime. The frozen seed gave \(\sigma_a^{-1/b}\) as its proportional life expression; that has the wrong sign. The correct stress exponent is \(1/b<0\).[1][2]
Basquin's 1910 paper did not use today's \(\sigma_f'\)/$2N_f$ notation. It examined stress-versus-repetition curves from endurance tests and observed that log coordinates made many bounded portions approximately straight. The law is therefore a fitted stress–life relationship, not a microscopic explanation of crack initiation, a guarantee of one exponent for every material, or a promise of infinite extrapolation.[3]
Structural Signature¶
Sig role-phrases: defined cyclic stress measure — cycles or reversals to failure — negative-exponent power segment — condition-specific coefficients — bounded fatigue regime and censored tests.
- Stress convention. The modern expression uses amplitude in a stated loading condition, commonly fully reversed. The original paper sometimes used maximum stress and repetitions. Swapping amplitude, range or maximum stress without recalibration changes the intercept and possibly the interpretation.[3][1]
- Life endpoint. \(N_f\) means cycles to a defined failure criterion; $2N_f$ counts reversals, not twice the number of independent test specimens. A specimen that has not failed by the test stop is a runout, not an observed infinite life.[3][2]
- Power segment. Taking logarithms gives \(\log\sigma_a=\log\sigma_f'+b\log(2N_f)\). Approximate straightness is a range-specific empirical finding, not a theorem about every stress scale.[3]
- Fitted constants. Material, geometry, stress ratio, processing, environment and test method can change intercept or slope. Composite data make stress ratio and orientation dependence especially visible.[4]
- Regime. Stress-life Basquin behavior is most naturally used where elastic strain dominates. Substantial cyclic plasticity needs a distinct plastic-strain term; a knee or endurance-limit behavior may break one log–log line.[1][3]
What It Is Not¶
Basquin's law is not fatigue itself. It correlates a chosen repeated-stress measure with life; it does not resolve crack nucleation, propagation, material microstructure or a sequence of cumulative damage states. It is not the Coffin–Manson plastic strain-life term. The combined strain-life relation uses Basquin's elastic term plus a plastic term, so identifying that full sum as “Basquin's law” would erase a necessary regime distinction.[1]
It is not a Miner damage-accumulation rule for a varying load spectrum. A Basquin S–N relation may provide constant-amplitude lives used by a separate accumulation model, but the sum-of-damage assumption does not follow from the Basquin power law. Nor is it the Paris law, which relates crack-growth rate to stress-intensity range, not total cycles to failure for a calibrated stress amplitude.[1]
Scope of Application¶
The relation supports interpolation, and cautiously bounded prediction, for a defined material/specimen and loading condition where fatigue-test points form an approximately linear log–log stress–life segment. Basquin's own wrought-iron bending example showed several failed tests fitting a line, but also lower-stress specimens that had not failed when the test ended. Those runouts warn against declaring that the same line extends through all very long lives.[3]
The form can be fitted outside historical metals: an original composite-laminate study writes \(\sigma^mN=c\) and uses it on glass/polyester and glass/epoxy fatigue datasets. It also shows that stress ratio, fiber orientation and frequency matter. The transfer is the power-law fit, not a claim that a wrought-iron slope applies to a fiber composite or even to a different laminate orientation.[4]
Clarity¶
Two plotting conventions can obscure the same relation. If log stress is plotted vertically and log life horizontally, the slope is \(b<0\) in the modern equation. If life is expressed as a power of stress, the exponent is \(1/b<0\), with an additional factor \(1/2\) in the reversals convention. A formula \(N_f\propto\sigma_a^{-1/b}\) would make life increase with stress when \(b<0\) and is not equivalent. Always state what the axes and signs mean.[1]
Likewise, $2N_f$ is a count of reversals; \(N_f\) is cycles. Replacing $2N_f$ by \(N_f\) while retaining the same \(\sigma_f'\) changes the intercept because \(2^b\) is not one. A fatigue-strength exponent or coefficient is a calibrated parameter, not a material constant independent of the test conditions.[2][4]
Manages Complexity¶
The straight-line log representation condenses many constant-amplitude fatigue tests into an intercept and a slope. It lets an analyst compare estimated life sensitivity to stress amplitude within a tested regime: with \(b=-0.1\), the inverted exponent would be \(-10\), illustrating how a modest stress change can strongly affect life if that fitted slope applies. This is an illustrative algebraic calculation, not a universal exponent or design value.[1]
The simplicity hides scatter, runouts, geometry, stress ratio, orientation and regime transitions. Basquin's original discussion noted tests that depart at low stress, and composite modeling explicitly tracks changes with condition. The useful compression is to organize a bounded dataset and identify where more testing is needed, not to replace uncertainty or specimen-specific validation.[3][4]
Abstract Reasoning¶
An empirical power segment \(\sigma=C N^b\) becomes a line \(\log\sigma=\log C+b\log N\). The negative \(b\) encodes an inverse stress–life relation in that segment. Re-expression in modern reversal notation changes \(C\) but preserves the slope if the same data and stress convention are used. Conversely, a changed slope signals that one straight line no longer summarizes the region or the loading condition.[3][1]
The relation's logic is conditional. To infer life at a new stress, the stress metric, failure endpoint, material state, mean-stress ratio, frequency and geometry should remain sufficiently comparable; uncertainty and censored runouts should be handled rather than discarded as failures. If macroscopic plastic strain dominates, the elastic stress-life segment is no longer the whole strain-life picture.[1][4]
Knowledge Transfer¶
The same fitting skeleton—repeated-stress magnitude, cycles to failure, log–log slope—appears in historical wrought-iron bending data and in composite-laminate fatigue analyses. A transferable question is whether the chosen test series really supports an approximately straight stress–life segment and which conditions determine its constants. This is more precise than transferring numerical slopes or physical failure mechanisms between materials.[3][4]
The generic power-law form also appears far beyond fatigue, but that is not sufficient to make Basquin's law a substrate-independent prime. Its constitutive variables remain cyclic mechanical stress and fatigue failure; the broader scaling idea belongs to separate prime review.[3]
Examples¶
Basquin's wrought-iron bending tests. His 1910 Curve A presents wrought-iron bars repeatedly bent and released. Several stress/repetition-to-rupture points lie near a straight line on log coordinates. Two lower-stress specimens were not observed to fail; one survived 48.2 million repetitions, making extrapolation visibly nontrivial.[3] Mapped back: cyclic stress measure = original maximum bending stress; life count = repetitions to rupture or censored test stop; power segment = Curve A fitted log line; coefficients = its fitted intercept and slope for the material/test form; boundary = low-stress runouts and limited observation range.
Composite laminate comparison. Ma and colleagues use \(\sigma^mN=c\), equivalently \(\log\sigma=p+q\log N\) with \(q=-1/m\), to fit glass/polyester and S2/glass/epoxy laminate fatigue datasets. Their comparisons make stress ratio and layup effects explicit rather than treating one fit as universal across all composites.[4] Mapped back: stress measure = cyclic applied stress under a specified ratio; life count = measured cycles to composite failure; power segment = fitted \(\sigma^mN=c\) line; coefficients = condition-specific \(m,c\); boundary = orientation, ratio and frequency dependence.
Structural Tensions¶
Compact interpolation versus regime-sensitive extrapolation. A single log–log line makes stress-life interpolation economical, but carrying that slope beyond failures into a possible endurance limit or into plasticity-dominated behavior can mislead. Restricting the line protects validity yet reduces prediction reach.[3][1] Diagnostic: Do failed observations and censored runouts support one slope across the intended stress and life range, or is there evidence of a knee or changed regime?
Pooled calibration versus condition-specific validity. Pooling materials or load conditions increases apparent data volume and simplifies one predictive equation, but stress ratio, geometry and composite fiber orientation can change the fitted relation. Separate condition-specific fits cost more testing and may leave wider uncertainty, while avoiding an unjustified shared slope.[4] Diagnostic: Are the test and target conditions equivalent enough to share coefficients, or must they be stratified or recalibrated?
Structural–Framed Character¶
Basquin's law is structural within a strongly domain-framed setting. Its power-law/log-line form travels between metallic and composite fatigue, but the stress, cycle and fatigue-endpoint roles are not optional decorations. It therefore lies nearer the domain-framed end than a bare mathematical power law.[3][4]
The five character criteria make that placement precise. Vocabulary travel: stress amplitude, cycles to failure and S–N slope transfer literally among fatigue tests, including metals and composites, but become metaphors in unrelated settings; only the bare power-law vocabulary travels without the materials frame. Evaluative weight: the equation itself is descriptive rather than a rule about acceptable risk or service life; safety margins and design decisions are applied judgments beyond its fitted relation. Human-practice dependence: fatigue under repeated load can occur without an engineer, although choosing specimens, failure endpoints and calibration conventions is a human measurement practice. Institutional origin: Basquin's name and ASTM-era endurance testing mark its history, not a membership requirement; an independently observed matching stress–life segment is still the same law. Import versus recognition: using the relation for composite tests is recognition when data and condition-specific fits support a power segment, while imposing the wrought-iron slope on a new material or across a runout is mere import.[3][1][4]
Its character: a reusable engineering stress–life abstraction with a strong structural equation and constitutive fatigue frame. A substrate-neutral inverse power-law skeleton is a future-prime question, not an already-admitted prime inferred from this one empirical law.
Structural Core vs. Domain Accent¶
The core is a bounded empirical power-law association between a defined cyclic stress and a defined fatigue-life count, with fitted constants and a log–log straight segment. Wrought iron, glass/polyester, rotating bending and laminate angle are accents. But mechanical fatigue and its stress/cycle metrics are essential; deleting them leaves only a generic inverse power law, not Basquin's law.[3][4]
Live Gradual Deterioration describes a broader process and live Allometry and Scaling Law is oriented to size-property scaling. Neither currently supplies a defensible strict DAG parent for this particular stress–life relation. A more precise fatigue-life or empirical-power-law parent could be considered later; the staged entry remains unparented rather than forcing a topical edge.
Instantiates / Related Primes¶
The relation uses a power-law shape and concerns fatigue deterioration, but relatedness is not an automatic typed parent. The generic scaling skeleton could become a future-prime question if a prime is defined broadly enough to cover condition-specific inverse relations without falsely claiming size-scaling or qualitative scale transitions.[3]
Its S–N output can be an input to separate engineering calculations. That does not make variable-amplitude damage accumulation part of Basquin's law, and it does not establish a strict relation to an unreviewed damage-rule node.[1]
Neighborhood in Abstraction Space¶
Basquin's Law sits in a sparse region of the domain-specific corpus (81st percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Structural Mechanics & Materials (19 abstractions)
Nearest neighbors
- Material Ratcheting — 0.85
- Paris' Law — 0.83
- Permissible Stress Design — 0.83
- Plane Strain Compression Test — 0.81
- Aggregate Modulus — 0.81
Computed from structural-signature embeddings · 2026-10-08
Not to Be Confused With¶
- A universal fatigue law: slopes and ranges are calibrated; runouts or regime changes can invalidate a single line.[3]
- Coffin–Manson plastic strain-life: a distinct term for cyclic plastic deformation, combined with the elastic term only in a larger model.[1]
- Miner damage accumulation: a separate variable-amplitude summation assumption.
- Paris crack-growth relation: growth rate versus stress-intensity range, not stress versus total life.
- A generic power law: mathematical resemblance alone omits the cyclic-stress and failure roles.
References¶
[1] D. J. Jones and P. Kurath, Cyclic Fatigue Damage Characteristics Observed for Simple Loadings Extended to Multiaxial Life Prediction, NASA Contractor Report 182126 (1988), §1.1.2, printed pp. 2–3. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m
[2] eFatigue, Fatigue Tests and Stress-Life (S–N) Approach, Basquin equation discussion of $2N_f$ reversals versus \(N_f\) cycles; direct PDF access intermittent. registry ↩a ↩b ↩c
[3] O. H. Basquin, “The Exponential Law of Endurance Tests”, Proceedings of the American Society for Testing Materials 10 (1910), 625–630, especially pp. 626–627; scanned mirror/OCR checked against ASTM volume contents. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q
[4] H. Ma et al., “Modeling the Effect of Stress Ratio, Loading Frequency and Fiber Orientation on the Fatigue Response of Composite Materials”, Polymers 14 (2022), §2 eqs. (1)–(3), Table 1, Figure 1 and §5. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k