High-cycle fatigue (HCF) testing evaluates how advanced structural materials withstand millions of repeated elastic stress cycles, producing the S-N curves and fatigue limits that underpin safe-life and fail-safe design. Accredited third-party laboratories perform these tests in accordance with ASTM, ISO, EN, GB, and JIS standards to support qualification of aerospace, automotive, energy, and medical components.
Technical Overview
High-cycle fatigue (HCF) describes fatigue failure that occurs when a material is subjected to cyclic stresses that remain predominantly within the elastic range, typically at stress amplitudes below the yield strength and at lives exceeding approximately 10⁴ to 10⁵ cycles. Because the applied strains are small, HCF testing is conducted under force (stress) control, which distinguishes it from strain-controlled low-cycle fatigue testing covered by ASTM E606 and ISO 12106. The central output of an HCF test campaign is the stress-life (S-N) curve, also known as the Wöhler curve, which correlates applied stress amplitude with the number of cycles to failure or to a defined runout condition.
For many steels and titanium alloys, the S-N curve approaches a horizontal asymptote known as the fatigue limit (or endurance limit), below which the material can theoretically survive an unlimited number of cycles. Face-centered cubic alloys such as aluminum and many nickel-based superalloys exhibit no true fatigue limit and instead fail at progressively lower stresses even beyond 10⁸ cycles, a phenomenon addressed by very-high-cycle fatigue (VHCF) studies in which cracks often initiate subsurface at inclusions or internal defects, producing characteristic "fish-eye" fracture features.
Key technical concepts applied throughout HCF testing include:
- S-N (Wöhler) curve construction and fatigue strength at a fixed life (e.g., 10⁶ or 10⁷ cycles)
- Stress amplitude (σa), mean stress (σm), and stress ratio (R = σmin / σmax), including fully reversed loading (R = −1)
- Runout criteria, typically 10⁷ cycles for steels and 10⁸–10⁹ cycles for VHCF programs
- Stress concentration effects at notches (Kt) and the influence of surface finish and residual stress
- Mean stress correction models (Goodman, Gerber, Walker) and Haigh diagram representation
- Statistical scatter characterization using log-normal and Weibull distributions
Testing Samples
High-Performance Metallic Alloys
Testing programs commonly cover high-strength low-alloy and maraging steels (e.g., 300M, 4340, 17-4PH), aerospace-grade aluminum alloys (2xxx and 7xxx series), titanium alloys such as Ti-6Al-4V, nickel-based superalloys, magnesium alloys, and stainless steels. These materials form the load-bearing backbone of flight structures, drivetrains, and rotating machinery, where fatigue performance is often the design-limiting property.
Additively Manufactured Materials
Laser and electron-beam powder bed fusion materials, including Ti-6Al-4V, AlSi10Mg, and 316L stainless steel, are increasingly subjected to HCF qualification. Test programs frequently compare as-built versus machined surfaces, heat-treated versus hot isostatically pressed (HIP) conditions, and different build orientations, because gas porosity, lack-of-fusion defects, and surface roughness can dramatically reduce fatigue strength relative to wrought equivalents.
Polymer Matrix Composites and Joints
Carbon and glass fiber-reinforced polymer laminates, sandwich structures, adhesively bonded joints, and mechanically fastened joints are tested under tension-tension or tension-compression cyclic loading. Composite fatigue behavior is governed by progressive damage accumulation (matrix cracking, delamination, fiber breakage) rather than a single dominant crack, following methods such as ASTM D3479 and ISO 13003.
Specimen Geometry and Preparation
Standard specimen types include cylindrical and flat hourglass profiles for uniform-stress testing, straight-sided gauges for composite coupons, and circumferentially notched specimens with defined Kt values (typically 1.5 to 3.0) for severity evaluation. ASTM E466 recommends a polished surface finish (for example, Ra ≤ 0.2 μm on metallic gauge sections) to minimize spurious surface initiation, and machining residual stresses must be controlled or documented. Bend bars and rotating-beam specimens follow the geometry requirements of the respective bending standards.
Testing Items
Accredited third-party laboratories offer a comprehensive portfolio of high-cycle fatigue test items tailored to material qualification and component verification:
- Axial stress-controlled fatigue (tension-compression, tension-tension) at selected stress ratios such as R = −1, 0.1, and 0.5
- Rotating beam (rotating bending) fatigue testing for rapid fatigue limit screening
- Three-point and four-point plane bending fatigue
- Torsional fatigue and combined axial-torsional high-cycle loading
- Full S-N (Wöhler) curve generation across the 10⁴ to 10⁷ cycle regime
- Fatigue limit and conditional endurance strength determination using the staircase method
- Notched fatigue strength testing with controlled stress concentration factors
- Very-high-cycle fatigue (VHCF) testing to 10⁸–10⁹ cycles using resonant or ultrasonic techniques
- Elevated-temperature fatigue testing in air or inert atmospheres
- Corrosion fatigue testing in aqueous, saline, or simulated physiological media
- Fretting fatigue evaluation for interference-fit and clamped interfaces
- Post-test fractographic analysis using scanning electron microscopy (SEM)
Testing Methods
Axial Force-Controlled Fatigue Testing
Axial HCF testing is performed to ASTM E466, ISO 1099, and GB/T 3075. Sinusoidal cyclic loads are applied to aligned specimens using servo-hydraulic or electromechanical dynamic testing machines at frequencies typically between 5 and 100 Hz. Bending introduced by misalignment is minimized and verified in accordance with ASTM E1012, and the dynamic force applied to the specimen is periodically verified per ASTM E467. Tests terminate at specimen failure, defined by separation or a specified drop in compliance, or at runout (commonly 2 × 10⁶ to 10⁷ cycles).
Rotating Beam Fatigue Testing
Rotating bending methods per ASTM E2948, ISO 1143, GB/T 4337, and JIS Z 2274 apply a fully reversed (R = −1) bending stress to a rotating specimen. Because each revolution constitutes one complete stress cycle, rotating beam systems reach high effective frequencies economically and are widely used to establish fatigue limits for bar, wire, and fastener materials.
Bending and Torsional Fatigue Methods
Plane bending and three-/four-point bending tests evaluate sheet materials and weldments, while torsional fatigue of thin-walled tubes follows ASTM E2207 to characterize shear-dominated crack growth. Combined axial-torsional loading enables investigation of multiaxial stress states representative of shafts and structural details under service spectra.
Statistical Design and Data Evaluation
Fatigue data exhibit substantial scatter, so test matrices are designed and analyzed statistically. Guidance in ASTM E739 governs the analysis of linearized S-N data, the staircase (Dixon-Mood) method estimates the fatigue limit at 50% survival probability, and Weibull statistics quantify life dispersion. Exploratory S-N curves typically employ 8 to 15 specimens, while statistically rigorous fatigue limit determination with defined confidence bounds may require additional specimens.
Ultrasonic and Resonant VHCF Testing
Ultrasonic fatigue systems excite specimens at approximately 20 kHz, allowing 10⁹ cycles to be accumulated within days. Because self-heating strongly influences results at such strain rates, temperature monitoring, cooling air, and pulse-pause control are mandatory, and results are interpreted alongside conventional-frequency reference data.
Failure Analysis and Reporting
Failed specimens are examined fractographically to identify initiation sites (surface inclusions, porosity, machining marks), stable crack propagation zones with beach marks and fatigue striations, and final overload fracture areas. Results are reported in accordance with ASTM E468, including material condition, specimen geometry, stress parameters, frequency, environment, cycles to failure or runout, and fitted S-N curves with survival probability statements.
Applications
High-cycle fatigue data directly govern design allowables, inspection intervals, and service life predictions across safety-critical industries:
- Aerospace: qualification of landing gear steels, engine disks and blades, wing attachments, fasteners, and helicopter dynamic components against flight-load spectra
- Automotive and rail: durability validation of crankshafts, connecting rods, valve springs, suspension arms, wheel hubs, bogie structures, and axles
- Power generation: life assessment of gas turbine hot-section superalloys, steam turbine shafts and blades, wind turbine drivetrain components, and nuclear structural materials
- Medical devices: fatigue verification of orthopedic implants and spinal fixation constructs to standards such as ASTM F382, ASTM F1717, and ISO 7206
- Oil, gas, and marine: corrosion fatigue evaluation of risers, mooring chains, weldments, and subsea hardware in seawater environments
- Additive manufacturing qualification: process parameter validation, post-treatment optimization, and supply chain certification of AM structural parts
FAQ
How does high-cycle fatigue differ from low-cycle fatigue testing?
High-cycle fatigue operates under force control with nominally elastic stresses at lives above roughly 10⁴ cycles, whereas low-cycle fatigue uses strain control with significant plasticity at shorter lives, following ASTM E606 and ISO 12106. The two regimes require different equipment control modes and produce different design data: S-N curves for HCF and strain-life (Coffin-Manson) relationships for LCF.
How many specimens are required to build an S-N curve?
An exploratory S-N curve generally requires 8 to 15 specimens distributed across four to six stress levels, consistent with ASTM E739. Determining a statistically defensible fatigue limit with the staircase method, including stated confidence intervals, typically requires a larger sample of 15 to 30 specimens.
What runout cycle count is used?
A runout of 10⁷ cycles is conventional for steels and titanium alloys, while aerospace programs frequently specify 10⁶ cycles as a design reference. For aluminum alloys and VHCF studies that exhibit continued life-dependent strength degradation, runouts of 10⁸ or even 10⁹ cycles are specified.
Does test frequency affect the results?
For most structural metals at moderate temperatures, frequency effects are negligible between 1 Hz and approximately 150 Hz. Significant effects arise for strain-rate-sensitive alloys, at elevated temperatures, in corrosive media (where time-dependent environmental interaction dominates), and in polymer matrix composites prone to hysteretic self-heating, so frequency selection must be justified for each material system.
Can testing be performed at elevated temperature or in corrosive environments?
Yes. Laboratory systems are equipped with environmental chambers and induction or resistance heating for testing in air, inert gas, or controlled atmospheres at temperatures relevant to engine applications, as well as recirculating cells for saline (e.g., 3.5% NaCl) or simulated body fluid corrosion fatigue testing.
How are results reported and used in design?
Deliverables include tabulated raw data, plotted S-N curves with fitted statistical bounds, fatigue limit estimates at defined survival probabilities, Haigh diagram data for mean stress effects, and fractographic findings identifying failure origins. Design engineers integrate these allowables with load spectra, damage-tolerant analysis, and safety factors to set inspection intervals and service lives.