Spring fatigue testing standard determination establishes the controlled procedures used to verify that compression, extension, torsion, and disc springs can endure millions of cyclic loading cycles without failure. Guided by internationally recognized standards such as ISO 1099, ASTM E466, GB/T 16947, and the EN 13906 series, accredited third-party laboratories measure fatigue life, S-N (Wöhler) curves, endurance limits, and load relaxation to support design verification, quality control, and product certification across automotive, aerospace, railway, and industrial applications.
Technical Overview
Springs are elastic machine elements designed to store and release energy, and in most service environments they operate through thousands to millions of repeated load cycles. Under cyclic loading, fatigue—not static overload—is the dominant failure mechanism. Microscopic cracks typically initiate at surface imperfections, non-metallic inclusions, corrosion pits, or damaged shot-peened layers on the wire surface, then propagate incrementally until sudden, often brittle fracture occurs at stress levels far below the material's ultimate tensile strength.
Because fatigue performance cannot be reliably inferred from static properties such as tensile strength or hardness, it must be measured directly under controlled cyclic conditions. Standard determination provides this framework by defining specimen requirements, loading parameters, environmental conditions, termination criteria, and statistical evaluation methods so that results are repeatable and comparable between laboratories. The most widely referenced frameworks include ISO 1099 and ASTM E466 for axial force-controlled fatigue testing of metallic materials, GB/T 3075 for the axial force-controlled fatigue test method, GB/T 16947 for fatigue testing of helical springs, the EN 13906 series for calculation and verification of cylindrical helical compression, extension, and torsion springs, and JIS B 2704 for helical spring design and testing requirements.
- Fatigue life: the number of cycles a spring sustains before fracture or rejection
- S-N curve (Wöhler curve): the relationship between cyclic stress amplitude and cycles to failure
- Endurance limit: the stress amplitude below which the spring survives a defined runout cycle count
- Load relaxation: progressive loss of spring force under sustained or cyclic loading
Testing Samples
Helical Compression Springs
The most frequently tested spring category, covering applications from precision instrument springs to heavy railway bogie springs. Specimens are cycled axially between specified deflection or force limits, and fatigue performance is evaluated in terms of cyclic torsional shear stress in the wire. Wire diameters may range from fractions of a millimeter to over 40 mm for rail and heavy industrial springs.
Helical Extension Springs
Extension springs are loaded in cyclic tension, and their fatigue life is frequently governed by the highly stressed hook, loop, or cross-body end configurations. Test programs document end geometry carefully, since a defect in the end formation can dominate the failure location and obscure the fatigue performance of the spring body.
Torsion Springs
Torsion springs are cycled through defined angular deflections, producing cyclic bending stress in the wire. Leg geometry, mandrel fit, and coil-to-coil contact must be controlled and reported, as they strongly influence the local stress distribution and resulting fatigue life.
Disc Springs (Belleville Springs)
Disc springs sustain very high loads over small deflections and are often tested in single or stacked configurations. Requirements for dimensional tolerances, material quality, and load testing are covered by standards such as EN 16983, and fatigue evaluation focuses on cyclic tensile stresses at the inner edge regions.
Wave, Die, Leaf, and Special-Purpose Springs
Wave springs, rectangular die springs, and leaf or flat springs are tested with fixtures that reproduce their actual installation and loading conditions. For leaf springs used in suspension systems, cyclic bending tests with realistic end constraints are typical.
Regardless of spring type, specimen preparation is critical for valid results:
- Springs must be representative of regular production, with no additional machining of active coils
- Lot traceability of wire material, coiling, heat treatment, and surface finishing (shot peening, coating, electropolishing) must be documented
- Free length, spring rate, and load at reference deflection are measured before testing to establish baselines
- S-N curve determination typically requires a sufficient number of springs distributed across multiple stress levels
Testing Items
Fatigue testing programs are structured around quantifiable performance items, each tied to acceptance criteria in the governing standard or the customer's product specification:
- Fatigue life determination (cycles to failure) at specified stress, deflection, or load levels
- S-N (Wöhler) curve construction across multiple stress amplitudes
- Endurance limit or fatigue strength at specified cycle counts such as 10⁵, 10⁶, or 10⁷ cycles
- Load loss and relaxation behavior under cyclic loading
- Permanent set and free length change after cycling
- Spring rate (stiffness) degradation over the test duration
- Post-test crack detection using dye penetrant or magnetic particle inspection
- Fracture surface (fractographic) analysis to confirm fatigue origin and failure mode
- Evaluation of surface treatment effects, including shot-peening residual stress benefits
Testing Methods
Test Equipment and Loading Control
Spring fatigue testing is performed on dedicated spring fatigue test machines or universal servo-hydraulic and electrodynamic systems equipped with precision load cells and displacement sensors. Following the force-controlled principles of ASTM E466 and ISO 1099, or the spring-specific procedures of GB/T 16947, the spring is cycled between an upper and lower load or deflection limit until fracture, excessive load loss, or the runout cycle count is reached.
Loading Parameters
The essential loading variables are mean load, load amplitude, and the stress ratio R (the ratio of minimum to maximum load or stress). Test frequency is selected to remain well below the spring's natural frequency, avoiding resonance and self-heating; coil temperature is monitored throughout, since elevated temperature accelerates relaxation and distorts fatigue results. Test stops at defined intervals permit intermediate measurement of free length and load.
S-N Curve Determination
A complete S-N curve is generated by testing groups of springs at several stress levels, typically spanning from a high amplitude producing failure within a few tens of thousands of cycles down to runout levels. Each stress level is populated with a statistically meaningful number of specimens, and the resulting data are fitted to produce the characteristic curve from which fatigue strength at a required survival probability is derived.
Load Relaxation and Permanent Set Measurement
Before and after cyclic testing, springs are removed and measured for free length and load at a reference deflection. Acceptance criteria commonly specify a maximum permissible load loss, for example a few percent of initial load, and a maximum free height change. These measurements are especially important for springs in safety assemblies where force retention is functionally critical.
Statistical Evaluation
Because fatigue data scatter inherently, standards require statistical treatment. Weibull distribution analysis with median-rank plotting is used to estimate survival probabilities at given stress levels, while the staircase (up-and-down) method may be applied to refine the endurance limit estimate. Results are typically reported at 90% or 95% survival probability with stated confidence levels.
Post-Test Failure Analysis
Fractured springs are examined to confirm the fatigue mechanism and identify the crack initiation site, distinguishing surface-origin fatigue from inclusion-initiated subsurface failure. Runout springs may be inspected non-destructively for incipient cracks, providing additional assurance that the runout result reflects true survival rather than undetected damage.
Applications
Spring fatigue test data directly support engineering decisions across nearly every industrial sector:
- Automotive: suspension coil springs, engine valve springs, clutch and brake return springs, fuel injection springs, and seat springs requiring validated life under road-load spectra
- Railway: bogie suspension and damping springs subject to millions of load cycles over service life, tested to demanding endurance requirements
- Aerospace: actuation, latch, and landing gear springs where failure consequences are severe and statistical substantiation is mandatory
- Industrial machinery: die springs, pump and compressor valve springs, and safety relief valve springs where force retention protects equipment and personnel
- Medical devices: surgical instrument and implant springs requiring documented fatigue validation for regulatory submission
- Electronics and consumer products: battery contact springs, switch springs, and display hinge springs tested for extended cycling reliability
Beyond design verification, fatigue test results are used for supplier qualification, incoming lot acceptance, comparative material and process evaluation (such as quantifying shot-peening improvements), root-cause investigation of field failures, and evidence packages for certification by regulatory bodies and accredited third-party laboratories.
FAQ
What is spring fatigue testing standard determination?
It is the process of selecting and applying the appropriate published standards—such as ISO 1099, ASTM E466, GB/T 16947, GB/T 3075, EN 13906, or JIS B 2704—to define specimen conditions, loading parameters, termination criteria, and statistical evaluation methods for measuring spring fatigue performance in a reproducible way.
Which standard should be used for a given spring?
Material-level axial fatigue methods are covered by ISO 1099, ASTM E466, and GB/T 3075; helical spring-specific fatigue testing follows GB/T 16947; design and verification rules for compression, extension, and torsion springs are given in EN 13906 parts 1 through 3; disc springs are covered by EN 16983. Product-specific customer specifications frequently add acceptance limits on top of these methods.
How many cycles define a runout?
Runout criteria depend on the application. Common values range from 10⁵ to 10⁷ cycles; automotive suspension springs may be qualified at several hundred thousand cycles with a load-loss limit, while valve springs and similar high-cycle components often require 10⁷ cycles or more without failure.
Should the test be force-controlled or displacement-controlled?
Both modes are used. Displacement control reproduces installations where the spring operates between fixed positions, while force control aligns with the general metallic fatigue practices of ISO 1099 and ASTM E466. Because spring stiffness can change as fatigue damage accumulates, the two modes can yield different life results, so the governing standard or product specification should dictate the choice.
How many specimens are required?
For a full S-N curve, specimens are distributed across at least three to five stress levels with a statistically meaningful number of springs per level. Simple verification or qualification tests may use three to five springs per condition, while statistical endurance limit determination requires larger sample sizes.
What factors most influence spring fatigue life?
Wire surface quality, shot-peening intensity and coverage, material cleanliness, coiling and heat-treatment quality, design stress amplitude, mean stress, corrosive environment, and operating temperature are the dominant variables. Small improvements in surface condition often produce large gains in fatigue life.
How long does fatigue testing take?
Duration depends on frequency and target cycles: one million cycles at 10 Hz take roughly 28 hours, while high-cycle runout tests at lower frequencies for large springs may run for several days to weeks per specimen. Test plans are usually scheduled in parallel across multiple test stations to compress program timelines.