Material Fatigue Property Analysis: Testing Methods, Standards, and Industrial Applications

Material fatigue property analysis is a critical evaluation process that determines how metals, composites, polymers, and finished components behave under repeated cyclic loading, providing essential S-N curve data, fatigue limits, and crack growth behavior for design and qualification. By following recognized standards such as ASTM E466, ASTM E606, ASTM E647, ISO 1099, ISO 12106, and the corresponding GB series, engineers can predict service life, prevent catastrophic failure, and optimize material selection. This article explains the principle, sample types, test items, methods, and typical applications of fatigue characterization in modern engineering.

Technical Overview

Fatigue is the progressive, localized, and permanent structural damage that occurs when a material is subjected to repeated or fluctuating stresses, often at levels well below the static yield or ultimate strength. Engineering statistics consistently indicate that the majority of mechanical service failures in load-bearing structures are fatigue-related, which makes material fatigue property analysis a cornerstone of structural integrity assessment, product qualification, and failure investigation.

Fatigue characterization is generally divided into two complementary approaches. The stress-life (S-N) approach evaluates the number of cycles to failure under controlled stress amplitudes and is used to establish the fatigue strength or endurance limit of a material, typically over ranges from 10,000 to 10 million cycles. The strain-life approach, applied mainly to low-cycle fatigue, measures the cyclic stress-strain response under controlled total strain amplitudes and supports elastic-plastic design. In addition, fracture-mechanics-based fatigue crack growth testing quantifies the rate of crack extension per cycle (da/dN) as a function of the stress intensity factor range, enabling damage tolerance analysis of structures that may contain manufacturing or service-induced flaws.

Key parameters investigated during fatigue analysis include stress amplitude, mean stress, stress ratio (R), loading frequency, surface condition, stress concentration, and environmental effects. Accredited third-party laboratories typically perform these evaluations in accordance with international standards such as ASTM E466, ASTM E606, ASTM E647, ISO 1099, ISO 12106, ISO 12107, ISO 1143, and the Chinese national standards GB/T 3075, GB/T 26077, GB/T 4337, and GB/T 6398, ensuring that results are traceable, statistically valid, and acceptable for certification purposes.

Testing Samples

Machined Metallic Coupons

Standard round or flat specimens are machined from bars, plates, castings, or forgings with carefully controlled geometry, surface roughness, and axiality in accordance with ASTM E466, ASTM E606, ISO 1099, ISO 12106, GB/T 3075, and GB/T 26077. Hourglass and uniform-gage profiles are the most common configurations for baseline material characterization.

Welded Joints

Transverse butt welds, fillet welds, and tubular joints are tested in the as-welded or post-weld heat-treated condition to reflect fabrication-related stress concentrations, residual stresses, and microstructural heterogeneity. Testing follows the applicable ISO and EN weld fatigue procedures and is essential for structures governed by fatigue design rules.

Fasteners and Threaded Components

Bolts, screws, studs, and other threaded fasteners are evaluated under axial tension-tension cycling as specified in ISO 3800, covering rolled, cut, and coated threads to verify preload retention and thread-root fatigue performance.

Actual Components and Subassemblies

Full-scale or representative parts such as shafts, springs, brackets, medical implants, and additively manufactured elements are tested to capture the influence of geometry, manufacturing process, surface treatment, and assembly conditions on real-world fatigue behavior.

Polymer and Composite Materials

Plastics and fiber-reinforced composites are tested using procedures such as ASTM D7774, ASTM D7791, ASTM D3479, and ISO 13003, with special attention to frequency-dependent self-heating and viscoelastic effects that require carefully controlled test rates.

Testing Items

• S-N curve (stress versus cycles-to-failure) determination for a specified survival probability
• Fatigue limit or endurance limit estimation using staircase or incremental step methods
• Fatigue strength at a specified number of cycles, such as 10,000, 100,000, or 10 million cycles
• Strain-controlled low-cycle fatigue testing with cyclic stress-strain and Manson-Coffin strain-life parameters
• Fatigue crack growth rate (da/dN) measurement and Paris law coefficient determination
• Threshold stress intensity factor range for crack growth arrest assessment
• Mean stress sensitivity evaluation and construction of Haigh or modified Goodman diagrams
• Effect evaluation of surface treatments, coatings, shot peening, and machining on fatigue strength
• Elevated-temperature, corrosion fatigue, and environmental condition testing
• Post-test fractographic examination to identify crack initiation sites and failure mechanisms

Testing Methods

Axial Load-Controlled High-Cycle Fatigue

Servohydraulic or electromechanical test systems apply constant-amplitude axial loading in tension-tension or tension-compression modes at stress ratios typically ranging from -1 to 0.5. The method is performed in accordance with ASTM E466, ISO 1099, and GB/T 3075, and is the primary technique for generating S-N data. Load verification is conducted following ASTM E467.

Strain-Controlled Low-Cycle Fatigue

A calibrated extensometer controls total strain amplitude while the system records the cyclic hardening or softening response. Testing follows ASTM E606, ISO 12106, and GB/T 26077 and produces the cyclic stress-strain curve and elastic-plastic strain-life parameters required for thermomechanical and high-strain design assessment.

Bending and Rotating Bending Fatigue

Rotating beam systems per ISO 1143 and GB/T 4337, together with plane bending and flexural methods per ASTM D7791, generate fatigue data for shaft-like components, wires, sheet materials, and polymers where bending stress dominates the service condition.

Fatigue Crack Growth Testing

Compact tension or middle-tension specimens containing a precrack are cycled under controlled stress intensity ranges, and crack length is monitored using compliance methods or optical measurement. Data reduction follows ASTM E647, ISO 12108, and GB/T 6398, providing da/dN curves, threshold values, and fracture-critical design inputs.

Statistical Evaluation and Presentation

Specimen allocation and result analysis are planned in accordance with ISO 12107 and ASTM E468, using statistical methods to define S-N curves and fatigue limits at confidence and probability levels appropriate for design allowables. Testing frequency, waveform, temperature, humidity, and environmental exposure are documented and reported with full traceability.

Applications

• Aerospace: airframe structures, landing gear, and engine discs, where low-cycle fatigue and damage tolerance verification are mandatory for certification
• Automotive: chassis components, suspension springs, welded body structures, and powertrain parts subjected to road-load spectra
• Rail transportation: axles, wheels, and bogie structures evaluated against EN and ISO railway material requirements
• Medical devices: orthopedic implants and instruments validated through fatigue endurance testing in simulated body environments
• Pressure equipment: vessels and piping subject to cyclic pressure and thermal loading, supporting fatigue assessment approaches referenced in EN design standards
• Oil and gas: drill strings, risers, and wellhead components requiring corrosion fatigue and crack growth data
• Civil infrastructure: bridge steels, welded connections, and cables under traffic and wind-induced cyclic loading
• Additive manufacturing: qualification of 3D-printed parts, where layer interfaces, porosity, and surface roughness strongly influence fatigue performance

FAQ

What is the difference between high-cycle and low-cycle fatigue?

High-cycle fatigue occurs at low strains and long lives, typically beyond 10,000 to 100,000 cycles, and is evaluated by stress-controlled S-N testing. Low-cycle fatigue involves significant plastic strains and short lives, usually fewer than about 100,000 cycles, and requires strain-controlled testing that captures elastic-plastic material behavior.

How many specimens are required to determine an S-N curve?

The specimen count depends on the statistical objective. As a general guideline guided by ISO 12107, a preliminary S-N curve may require 8 to 15 specimens, while staircase determination of the fatigue limit typically needs 15 to 30 specimens to achieve acceptable confidence levels for design allowables.

What test frequencies are used in fatigue testing?

Conventional servohydraulic testing is commonly performed between 5 Hz and 60 Hz, while resonance-based systems can reach several hundred hertz for long-life metallic testing. For polymers and composites, frequencies are deliberately limited to prevent hysteretic self-heating, and strain-rate sensitivity is taken into account.

Can fatigue testing be performed at elevated temperature or in corrosive media?

Yes. Environmental chambers and corrosion cells allow testing at sub-ambient to elevated temperatures and in controlled aqueous or sour environments, enabling evaluation of temperature-dependent fatigue strength and corrosion fatigue behavior relevant to power generation, chemical processing, and subsea applications.

Why is fatigue crack growth testing important?

Crack growth testing establishes how quickly an existing flaw extends under cyclic loading. Combined with inspection intervals and fracture toughness data, it supports damage tolerance analysis that determines whether a cracked component can remain safely in service until the next scheduled inspection.

How long does a fatigue test take?

Test duration depends on the target life and frequency. A 10-million-cycle run at 50 Hz requires approximately two to three days per specimen, while low-cycle and crack growth tests are usually completed within hours to a few days. Overall project timelines also depend on the number of stress levels and statistical plan defined at the quotation stage.