Extension spring fatigue performance testing verifies that tension springs can withstand millions of cyclic loading cycles without failure, crack initiation, or unacceptable load loss. This article explains the testing standards (ISO, ASTM, EN, GB, JIS), sample requirements, test items, methods, and evaluation criteria used by accredited third-party laboratories to qualify extension springs for safety-critical applications.
Technical Overview
Extension springs (also called tension springs) are helical coil springs designed to operate under axial tensile loading. Unlike compression springs, they are typically manufactured with closed coils and initial tension, and they absorb and store energy when extended by end hooks or loops. Because most extension springs in service are subjected to repeated, cyclic tensile loading rather than a single static load, fatigue performance is often the decisive design criterion.
Fatigue failure of an extension spring occurs through the progressive initiation and propagation of a crack under cyclic stress, usually at locations of stress concentration. The most common failure sites are the transition bends of the end hooks or loops, where localized bending stresses can exceed the torsional stress in the spring body by a significant margin, followed by the inner surface of the active coils where wire torsional stress peaks. Surface quality, decarburization, wire drawing marks, shot peening condition, hook geometry, and corrosive environment all strongly influence fatigue life.
Extension spring fatigue testing quantifies the relationship between cyclic load (or stress) and the number of cycles to failure (the S-N or Wöhler curve), determines the endurance limit, and measures load relaxation over the design life. Testing is typically performed in accordance with recognized frameworks such as ISO 1099 (metallic materials, axial force-controlled fatigue testing), ISO 12107 (statistical planning and analysis of fatigue data), ASTM E466 (force-controlled axial fatigue of metallic materials), ASTM A125 (fatigue testing of steel springs), EN 13906-2 (calculation and qualification of helical extension springs), GB/T 1239.1 and GB/T 16947 (Chinese national standards for cold-cooled tension springs and helical spring fatigue testing), and JIS B 2704 (helical compression and extension springs). Accredited third-party laboratories use these frameworks to generate traceable, statistically defensible fatigue data for design validation, supplier qualification, and failure analysis.
Testing Samples
Material Categories
Fatigue behavior depends fundamentally on the spring wire material and its processing history. Common sample categories submitted for testing include:
- Cold-drawn high-carbon steel wire (music wire, ASTM A228-type material) for general-purpose, high-strength springs.
- Oil-tempered chrome-silicon and chrome-vanadium alloy wire (ASTM A401 and ASTM A230-type materials) for elevated-strength and moderately high-temperature automotive and industrial springs.
- Austenitic stainless steel wire (302, 304, 316, ASTM A313-type material) for corrosion-resistant applications.
- Precipitation-hardening stainless steel wire (17-7 PH condition CH900) for aerospace and high-cycle-duty springs requiring corrosion resistance plus high fatigue strength.
- Copper alloys such as phosphor bronze and beryllium copper for electrical, marine, and spark-free applications.
End Configurations
Because the end configuration governs the dominant stress concentration, samples should be representative of production parts. Typical end configurations include machine half loops, full machine loops across the center, side loops, raised (extended) hooks, screwed-in hooks, and coned-end designs with short swivel hooks. The fatigue test report should always document the exact hook style, hook bend radius, and hook orientation relative to the spring axis.
Sample Preparation and Conditioning
Test springs must be in the final delivered condition: the same coiling, stress-relieving, shot peening (if specified), plating or passivation, and presetting operations as production parts. Machining, grinding, or re-bending of hooks after delivery is not permitted, since it alters the surface integrity that controls fatigue initiation. Before testing, each sample should be dimensionally characterized (wire diameter, mean coil diameter, free length, number of active coils, hook gap and angle) and load-tested to establish initial tension, spring rate, and load at test deflection. Surface condition should be verified visually and, where required, by penetrant inspection for circumferential wire defects. Sample quantities per load level are determined statistically, typically following ISO 12107 guidance; a minimum of 6 to 10 springs per stress level is common for S-N characterization, while fewer samples may suffice for pass/fail validation at a single load level.
Testing Items
The scope of an extension spring fatigue test program typically includes the following measurable items:
- Cycles to failure at specified maximum load and load ratio, reported as N (number of cycles) for each specimen.
- S-N (stress-cycle) curve construction across at least three load levels to characterize the fatigue strength of the spring population.
- Fatigue strength and endurance limit at specified survival probabilities (for example, 50% and 90% survival) at 10^5, 10^6, and 10^7 cycles.
- Load loss and permanent set: change in free length, load at reference deflection, and initial tension after cyclic loading.
- Spring rate degradation over accumulated cycles, measured by periodic static load checks.
- Failure mode classification: hook fracture, transition-bend fracture, body coil fracture, or no failure (runout).
- Fractographic examination of failed specimens to identify initiation sites, crack propagation characteristics, and contributory defects such as inclusions, seams, tooling marks, or hydrogen embrittlement from plating.
- Optional environmental fatigue evaluation under humidity, salt spray exposure, or elevated temperature where service conditions demand it.
Testing Methods
Constant-Amplitude Axial Fatigue Testing
The primary method is tension-tension cyclic loading on a dedicated spring fatigue testing machine. The spring is extended between fixed hook fixtures and cycled between a minimum load Fmin and a maximum load Fmax. The load ratio R = Fmin/Fmax is selected to reflect the service spectrum; values between 0.1 and 0.5 are typical for extension spring qualification because the spring must remain under net tension throughout the cycle. Load levels are commonly expressed as percentages of the maximum working load or as computed peak stresses in the wire body and hook bends. Testing continues until specimen failure or until the specified runout count is reached (typically 10^6 cycles for automotive chassis components, up to 10^7 cycles for long-life or aerospace applications).
S-N Curve Determination and Statistical Evaluation
To establish a full fatigue characterization, groups of springs are tested at three or more load levels spanning the expected failure range. Results are plotted as maximum stress (or load) versus log cycles to failure, and the data are evaluated statistically per ISO 12107 to derive the mean S-N curve, the fatigue strength at defined survival probabilities, and confidence intervals. The staircase (up-and-down) method may be applied near the endurance limit to refine the fatigue limit estimate with economical sample counts.
Load Relaxation and Permanent Set Measurement
Companion specimens are cycled to intermediate cycle counts (for example, 10^4, 10^5, and 10^6 cycles) and removed for static characterization. Free length, load at reference deflection, initial tension, and spring rate are re-measured and compared with the pre-test baseline to quantify relaxation. Acceptance is commonly defined as a load loss not exceeding 5% or a free length change within specified drawing tolerances, unless the application standard dictates otherwise.
Equipment, Fixtures, and Test Control
Suitable equipment includes servo-hydraulic fatigue frames, electromagnetic dynamic actuators, and cam- or crank-driven spring fatigue testers designed for high-frequency production testing. Fixtures must grip the hooks without introducing artificial stress raisers or bending moments; loading axis alignment is verified to prevent off-axis loading that would corrupt results. Test frequency is selected to avoid self-heating of the wire (typically below approximately 10 to 20 Hz for small springs, and lower for large-section wire), and laboratory ambient conditions (temperature, humidity) are recorded. Load cell calibration and periodic load verification are maintained traceable to national standards, and all specimens are inspected continuously by fracture detection circuitry or displacement monitoring so that the exact failure cycle is captured.
Post-Test Evaluation
Failed springs are photographed, fracture surfaces are cleaned and examined using stereo microscopy and, where necessary, scanning electron microscopy to distinguish fatigue striations from overload features and to identify initiation defects. Runout springs are dimensionally re-inspected and statically load-tested to quantify degradation. The final test report presents raw cycle counts, statistical analysis, S-N curves, failure mode distribution, relaxation data, and conclusions relative to the acceptance criteria.
Applications
Extension spring fatigue testing supports qualification and quality control across a wide range of industries:
- Automotive: hood, trunk, and door counterbalance springs, brake and clutch pedal return springs, throttle and shift return springs, seat and belt retractor mechanisms, and engine bay actuation springs.
- Aerospace: control system return springs, latching and detent springs, access-panel springs, and cargo restraint hardware, where demonstrated fatigue life at high survival probability is mandatory.
- Industrial machinery: valve return springs, counterbalance springs for machine guards and doors, textile and packaging equipment springs, agricultural implement springs, and conveyor tensioning springs.
- Consumer products: garage door and gate extension springs, trampoline springs, fitness equipment, retractable cord mechanisms, and appliance components where sudden fatigue failure creates safety risk.
- Medical devices: surgical instrument return springs, drug delivery and diagnostic device mechanisms requiring assured lifetime reliability.
- Electronics and electrical equipment: switch return springs, connector retention springs, and relay mechanisms, including conductive copper-alloy spring applications.
In each of these sectors, fatigue test data are used for design validation, supplier comparison, incoming lot qualification, root-cause investigation of field failures, and documentation for regulatory or customer approval.
FAQ
How many cycles should an extension spring withstand before failing?
The required life depends entirely on the application. Light-duty consumer springs may be qualified at 5,000 to 50,000 cycles, automotive components commonly require 10^5 to 10^6 cycles, and aerospace or continuously cycled industrial springs are often validated to 10^7 cycle runout. The specification should state both the load conditions and the required survival probability, not simply a cycle count.
Why do extension springs usually fail at the hooks rather than in the coils?
The end hooks experience combined bending and torsion at a sharp geometric transition, and the local stress at the inner bend radius can be substantially higher than the body coil stress, often estimated at up to 1.5 times or more depending on bend radius. Unless the hook radius is generous and the surface is defect-free, crack initiation occurs there first. Fatigue testing confirms whether the hook design or the body governs the life of a given spring.
What load ratio and load levels are used in testing?
Because extension springs cannot sustain compression, testing is performed under tension-tension cycling with a load ratio R between roughly 0.1 and 0.5, chosen to represent the actual service duty cycle. Load levels for S-N characterization are selected so that the lowest level approaches the expected endurance limit and the highest level produces failure within a practical number of cycles, typically spanning 10^4 to 10^7 cycles across the test matrix.
Does shot peening or plating affect the test program?
Yes. Shot peening induces compressive residual stress that measurably raises fatigue strength, so specimens must reflect the exact production peening condition. Electroplating, conversely, can introduce hydrogen absorption and reduce fatigue life or cause delayed brittle fracture, so plated springs should be tested in the delivered plated and baked condition to capture the true performance.
How many springs are needed for a statistically valid fatigue test?
For a pass/fail validation at a single load level, a small group such as 4 to 8 specimens is often acceptable. For full S-N curve and endurance limit determination with stated confidence, ISO 12107-based planning typically requires 6 to 10 specimens per load level across at least three levels. The exact plan should be agreed between the client and the accredited third-party laboratory before testing begins.
Can fatigue testing predict exact service life?
Fatigue data are inherently statistical. Testing establishes the probability of survival at given load levels rather than a deterministic life. By combining laboratory S-N data with the measured service load spectrum and appropriate safety factors, engineers can predict life with a defined confidence level, which is the accepted engineering basis for reliable spring design.