Spring fatigue characteristics determination is a specialized mechanical testing process that evaluates how compression, extension, torsion, disc, and leaf springs behave under repeated cyclic loading over thousands to millions of cycles. By quantifying fatigue life, endurance limits, load loss, and permanent set, manufacturers can validate spring designs, predict service life, and prevent premature failure in safety-critical automotive, aerospace, and industrial applications. This article explains the technical principles, sample requirements, test items, testing methods, and applicable standards involved in professional spring fatigue evaluation.
Technical Overview
Springs are elastic mechanical elements that store and release energy through repeated deflection. In service, they are almost always subjected to fluctuating stresses, and fatigue is the dominant failure mechanism: cracks initiate and propagate gradually at stress levels far below the static tensile strength of the material, eventually causing sudden fracture without warning. Determination of spring fatigue characteristics is therefore essential for any application where a spring must survive millions of operating cycles.
Fatigue characterization of springs typically establishes the relationship between cyclic stress (or cyclic deflection and load) and the number of cycles to failure, expressed as an S-N curve (also called a Wöhler curve), as well as the fatigue limit or endurance strength at a specified survival probability and cycle count. Closely related degradation phenomena are also measured, including load loss caused by stress relaxation and permanent set, which reduce spring performance even before fracture occurs.
Fatigue cracks in springs usually initiate at surface imperfections such as scratches, seams, non-metallic inclusions, corrosion pits, or decarburized layers. Key influencing factors include the wire material grade (e.g., cold-drawn carbon steel, chrome-silicon, chrome-vanadium, stainless steel, or nickel superalloys), spring index, wire diameter, surface finish, residual compressive stress introduced by shot peening, mean stress and stress ratio, working temperature, and corrosive media. A systematic fatigue testing program allows designers to account for all of these variables and to verify reliability using statistical tools such as Goodman diagrams and Weibull analysis.
Testing Samples
Test specimens should be actual production springs representative of the manufactured batch, processed under the same conditions as the delivered product, including coiling, heat treatment, presetting, and surface treatments such as shot peening or coating. Specimens must be free of handling damage, and free height, coil geometry, and surface condition should be verified before testing. Depending on the statistical method applied, a typical program uses at least 3 to 5 stress levels with 3 to 6 springs per level.
Helical Compression Springs
The most widely tested spring type, covering applications from engine valve springs and fuel injection springs to suspension coils and die springs. They are cycled in axial compression between parallel platens, with guiding to prevent buckling and lateral instability.
Helical Extension Springs
Extension springs are loaded in cyclic axial tension using hook or threaded-end fixtures. Particular attention is paid to the end loops and hooks, where bending stress concentrations make fatigue initiation likely.
Torsion Springs
Torsion springs are cycled in angular deflection between defined angular limits, with the wire subjected to fluctuating bending stress. Fixtures support the legs and control the mandrel or arbor over which the spring operates.
Disc Springs and Spring Washers
Belleville disc springs, whether tested individually or in stacks, are loaded axially between flat surfaces and cycled over a defined deflection range. Fatigue behavior depends strongly on the ratio of deflection to the flat position of the disc.
Leaf Springs, Wave Springs, and Special Designs
Automotive and commercial vehicle leaf springs are tested in cyclic bending on full-scale rigs, while wave springs, stamped flat springs, snap domes, battery contacts, and constant-force springs are tested with application-specific fixtures that replicate the actual installation geometry.
Testing Items
A comprehensive determination of spring fatigue characteristics may include the following evaluation items:
- Fatigue life: number of cycles to first visible crack or complete fracture at each stress or deflection level
- S-N (Wöhler) curve determination and regression fitting over the required stress range
- Fatigue limit or endurance strength at a specified cycle count and survival probability, commonly 106 to 107 cycles
- Fatigue strength verification testing at a specified load level and target cycle count
- Load loss (force relaxation) after cyclic loading, expressed as a percentage of initial load
- Permanent set and free length change after testing
- Spring rate (stiffness) variation before and after cyclic loading
- Mean stress and stress ratio influence, evaluated using Haigh or Goodman diagrams
- Failure mode and fractographic analysis: crack origin, propagation zone, beach marks, and final fast-fracture zone
- Optional environmental studies: elevated-temperature relaxation and corrosion fatigue combined with salt spray exposure
Testing Methods
Spring fatigue tests are performed on dedicated spring fatigue testing machines, servo-hydraulic or servo-electric universal testing machines, or high-frequency electrodynamic testers. The equipment is equipped with precision load cells, displacement sensors, automatic cycle counters, and fracture detection systems that shut down the machine when the load suddenly drops, indicating crack initiation or breakage.
Tests are most commonly conducted under deflection (strain) control, because most springs operate between fixed positions in service; force-controlled testing is used when the duty cycle is defined by fluctuating forces. The test defines a mean deflection or mean load plus a constant amplitude, or equivalently a stress ratio R (minimum stress divided by maximum stress). Test frequencies typically range from 1 Hz to 50 Hz depending on the machine type and specimen size, chosen high enough to accelerate testing while avoiding self-heating of the spring.
A standard test sequence proceeds as follows:
- Pre-test characterization: measurement of free height, outside diameter, wire diameter, and surface condition; determination of the initial load-deflection curve and spring rate
- Specimen mounting: careful centering and alignment, guiding against buckling, and consistent lubrication or dry conditions matching the intended service
- Cyclic loading: application of the selected mean load and amplitude at each stress level, running continuously without interruption
- In-process monitoring: continuous recording of peak and valley loads; any decline in force indicates cracking and is logged with the corresponding cycle count
- Runout termination: if no failure occurs, testing stops at the predetermined runout cycle count, commonly 107 cycles
- Post-test evaluation: measurement of permanent set and free height, recalculation of spring rate, quantification of load loss, and microscopic or fractographic examination of failed specimens
Data evaluation includes median and mean fatigue life at each level, statistical S-N curve regression, staircase or probit methods for fatigue limit determination, and reliability analysis using log-normal or Weibull distributions. Tests are commonly performed in accordance with recognized standards and guidelines, including ASTM E466 for force-controlled axial fatigue of metallic materials, ASTM E468 for reporting fatigue test results, ASTM E739 for statistical analysis of S-N data, GB/T 16947 for fatigue testing of helical springs, the EN 13906 series for design and verification of cylindrical helical springs, and the relevant JIS B 2700 series and SAE spring design guidelines. ISO/IEC 17025-accredited third-party laboratories can conduct these programs and issue internationally recognized test reports.
Applications
Spring fatigue testing supports product development, quality control, supplier qualification, and failure analysis across virtually every engineering industry:
- Automotive: engine valve springs, fuel injection and throttle return springs, clutch springs, brake springs, suspension coil and leaf springs, and seat belt retractor springs
- Rail transit: bogie axle-box springs, bolster springs, and coupling draft gear springs subjected to millions of load cycles
- Aerospace: landing gear springs, actuator and latch springs, and constant-force springs where failure is unacceptable
- Oil, gas, and petrochemical: pressure relief and safety valve springs requiring verified long-term reliability
- Industrial machinery: heavy-duty die springs, hydraulic valve springs, pump springs, and vibration isolation springs
- Electronics and electrical equipment: switch springs, connector contacts, battery contacts, and snap domes with high actuation counts
- Medical and pharmaceutical devices: surgical instrument springs, insulin pen springs, and drug delivery mechanisms
- Research and development: material grade comparison, shot peening and coating process validation, life prediction model calibration, and root cause analysis of field failures
FAQ
What is the difference between fatigue life testing and stress relaxation testing?
Fatigue life testing determines the number of cycles a spring can withstand under cyclic loading before cracking or fracture, whereas stress relaxation testing measures the gradual loss of spring force when the spring is held at constant deflection, often at elevated temperature. Both phenomena affect long-term durability, and a complete fatigue characterization program frequently includes both.
How many cycles are considered infinite life for a spring?
Most specifications treat 107 cycles without failure as the runout criterion representing practical infinite life, although some product standards accept 106 cycles for specific spring categories. The required runout count should always be agreed upon based on the applicable standard and the actual duty cycle of the application.
How many specimens are needed to establish an S-N curve?
A typical program tests 3 to 5 stress levels with 3 to 6 springs at each level, generally requiring 12 to 30 specimens in total. Determining the fatigue limit statistically with the staircase method requires an additional series of specimens, so the exact quantity depends on the desired confidence level.
Is the test controlled by force or by deflection?
Springs are usually tested under deflection control because they typically operate between fixed positions in service. Force control is selected when the application applies a controlled fluctuating load. The control mode must be specified in the test plan because it influences the measured life.
How long does spring fatigue testing take?
Duration depends on frequency and target cycles. At 30 Hz, 107 cycles require approximately four days of continuous running, while lower frequencies or higher runout counts extend the program to several weeks. High-frequency spring fatigue testers are often used to shorten project timelines.
What does the final test report include?
A professional report from an accredited third-party laboratory documents the sample description, applied standards and equipment, test conditions including mean load, amplitude, and frequency, individual cycles to failure, the fitted S-N curve and fatigue limit, load loss and permanent set results, fractographic findings, and clear pass or fail conclusions against the specified acceptance criteria.