Spring fatigue curve determination is a specialized mechanical testing process used to evaluate how compression, extension, torsion, and disc springs perform under repeated cyclic loading. By generating S-N (Wöhler) fatigue curves, accredited third-party laboratories help manufacturers validate spring designs, predict service life, and prevent premature failure in safety-critical components. This article explains the technical principles, sample types, test items, testing methods, standards, and typical industrial applications of spring fatigue curve measurement.
Technical Overview
Springs in service rarely fail under a single static overload. Instead, they degrade progressively under millions of repeated load cycles, and the dominant failure mechanism is metal fatigue. Spring fatigue curve determination, also known as S-N curve or Wöhler curve testing, quantifies the relationship between the cyclic stress applied to a spring and the number of cycles it sustains before fracture or unacceptable performance loss. The resulting curve is the foundation for fatigue life prediction, safe design stress selection, and durability validation.
During a fatigue test, the spring is cycled between a maximum and minimum deflection or load, producing a corresponding maximum stress, minimum stress, mean stress, stress amplitude, and stress ratio R (minimum stress divided by maximum stress). For helical springs the governing stress is torsional shear stress on the wire; for torsion and leaf springs it is bending stress; for disc springs it is a combined stress state. Test results are plotted as stress amplitude or maximum stress versus the number of cycles to failure on logarithmic coordinates, revealing a finite-life region at high stress and, for most spring steels, a plateau or endurance limit in the long-life region.
Failure is not limited to fracture. Cyclic loading can also cause spring set and load loss (relaxation), where the spring gradually loses free length and force without cracking. A complete fatigue evaluation therefore often combines S-N curve generation with load-loss measurement and fracture analysis. Relevant reference standards include GB/T 16947 for helical spring fatigue testing, the GB/T 1239 series for spring inspection, ASTM E466 and ASTM E468 for fatigue testing and reporting of metallic materials, ASTM E467 for dynamic force verification of fatigue testing machines, ASTM E739 for statistical analysis of S-N data, EN 13906-1 and EN 13906-2 for spring design verification, ISO 26910-1 for shot-peened springs, and JIS fatigue testing standards such as JIS Z 2273.
Testing Samples
Helical Compression Springs
Compression springs are the most frequently tested spring type, covering valve springs, clutch springs, suspension springs, die springs, pump springs, and safety valve springs. Testing is performed between defined deflection limits or load limits, and specimens are typically characterized for free length, spring rate, and load at test height before the fatigue program begins.
Extension Springs
Extension springs are cycled in tension between minimum and maximum elongation. Because the end hooks or loops are stress-concentration sites, fatigue failures often initiate at the hook bend, and special fixtures are required to apply pure axial loading without introducing bending artifacts.
Torsion Springs
Torsion springs and double torsion springs are cycled through defined angular deflections, producing alternating bending stress in the wire. Test fixtures clamp the spring legs and rotate them through precise angles while recording the torque and cycle count.
Disc (Belleville) Springs and Spring Stacks
Disc springs are fatigue tested individually or as stacked columns under axial cyclic deflection. Because fatigue cracks in disc springs typically initiate at the inner edge on the concave surface, testing and inspection focus on this critical region.
Special and Custom Springs
Wave springs, constant-force springs, wire forms, spiral springs, leaf springs, and stamped flat springs can also be evaluated. Common materials include carbon steel spring wire, oil-tempered wire, chromium-silicon and chromium-vanadium alloy wire per ASTM A401 and ASTM A232, cold-drawn steel music wire per ASTM A228, stainless spring steels per ASTM A313, and nickel-based or copper-based alloys for corrosive or high-temperature environments.
Testing Items
Depending on the engineering objective, a spring fatigue test program may include the following items:
- S-N (Wöhler) fatigue curve determination at one or more stress ratios
- Fatigue life verification at specified working stress and stroke
- Endurance limit and knee-point determination for the long-life region
- P-S-N curves with defined survival probability levels for statistical design
- Cyclic load loss and permanent set (relaxation) measurement during testing
- Dynamic spring rate and hysteresis evaluation under cyclic loading
- Mean stress sensitivity assessment and fatigue strength diagram construction
- Fracture surface analysis to identify crack initiation sites and failure mechanisms
- Effect evaluation of surface treatments such as shot peening, nitriding, or coating
Testing Methods
Specimen Preparation and Pre-Test Characterization
Specimens are taken from production lots or prototype batches and must be representative of the actual manufacturing process, including coiling, heat treatment, and surface finishing. Before cycling, each spring is measured for free length, outside diameter, squareness, permanent set, spring rate, and load at the specified test deflection, providing baseline data for detecting degradation during the test.
Equipment and Fixturing
Testing is performed on dedicated spring fatigue testing machines, including cam-type mechanical testers, servo-hydraulic systems, and electromagnetic or electrodynamic actuators. Machines are equipped with compression platens, tension hooks, or torsion fixtures matched to the spring geometry, and dynamic force verification is carried out in accordance with ASTM E467. Environmental chambers can be integrated for elevated-temperature, low-temperature, or corrosive-atmosphere fatigue evaluation.
Loading Procedure and Parameters
A full S-N curve typically requires four to six stress levels spanning the finite-life region, with a statistically meaningful number of springs tested at each level. Key controlled parameters include maximum and minimum deflection or load, stress ratio R, cycling frequency, waveform, and environmental conditions. The test frequency is selected well below the fundamental natural frequency of the spring, generally below one-tenth of it, to avoid spring surge, resonance, and self-heating that would distort the fatigue result.
Failure Criteria and Runout Definition
A specimen is defined as failed when complete fracture occurs, when a visible fatigue crack is detected, when permanent set exceeds the specified tolerance, or when load loss at the test height exceeds an agreed percentage such as five percent. Springs that survive the specified runout cycle count, commonly between one million and ten million cycles depending on the product specification, are suspended and treated as runouts in the data analysis.
Data Analysis and Curve Presentation
Failed and runout data are plotted on logarithmic stress-life coordinates and fitted using power-law regression in the finite-life region, consistent with ASTM E468 reporting practice and ASTM E739 statistical methodology. Where high reliability is required, the staircase method or distributed-stress testing is used to refine the endurance limit estimate, and Weibull analysis may be applied to derive probability-based P-S-N curves. The final report presents the fitted fatigue curve, runout points, individual data scatter, fracture photographs, and design recommendations for allowable working stress.
Applications
Fatigue curve data directly supports design decisions wherever springs operate under continuous cyclic loading. Typical applications served by accredited third-party laboratories include:
- Automotive engineering: engine valve springs, clutch and transmission springs, suspension springs, brake springs, and seat belt retractors
- Railway and commercial vehicles: bogie suspension springs, coupler springs, and pantograph components
- Aerospace: actuator springs, latch springs, and high-reliability fastener and control springs
- Industrial machinery: die springs, valve springs for hydraulic and pneumatic systems, pump springs, and seal springs
- Oil, gas, and power generation: safety relief valve springs and turbine-related spring components
- Medical and analytical devices: precision stainless springs in pumps, injectors, and surgical instruments
- Electronics and consumer products: battery contacts, switch springs, and snap-fit spring elements
Manufacturers use the resulting curves to select safe working stresses, set inspection and replacement intervals, qualify shot-peening and material substitutions, and demonstrate compliance with customer specifications and regulatory durability requirements.
FAQ
How many springs are needed to determine a fatigue curve?
A typical S-N curve program uses 15 to 30 springs, distributed across four to six stress levels with at least three specimens per level. More specimens are recommended when a probability-based P-S-N curve or a precise endurance limit is required, because fatigue data are inherently scattered.
What cycle count is used as the runout limit?
Runout is defined by the product specification and typically ranges from one million to ten million cycles. Ten million cycles is commonly applied for springs intended for effectively unlimited service life, while automotive chassis components are often evaluated at lower specified runout counts.
What testing frequency is used, and does frequency affect results?
Spring fatigue tests commonly run between 1 Hz and 30 Hz. The frequency must remain far below the natural frequency of the spring to prevent resonance and heating. Within this range, frequency effects on steel springs are usually minor, but elevated-temperature testing requires additional care to avoid artificial self-heating of the specimen.
How is failure defined if the spring does not break?
Non-fracture failure is defined by functional degradation, most often load loss at the test height beyond an agreed limit or permanent set beyond dimensional tolerance. These criteria are agreed with the client before testing because they depend on the application's functional requirements.
Can fatigue testing be performed at high temperature or in corrosive environments?
Yes. Environmental chambers allow fatigue curve determination at elevated or sub-zero temperatures, and salt-spray or humid atmospheres can be combined with cyclic loading to evaluate corrosion fatigue performance of stainless and coated springs.
What is the difference between a fatigue test and a relaxation test?
A fatigue test determines cycles to fracture under cyclic loading, producing an S-N curve. A relaxation or set test holds or cycles the spring under load and measures force decay and length change over time, focusing on load stability rather than crack initiation. Both assessments are frequently combined in a single test program.