Hardware spring fatigue failure analysis and life prediction technology
Fundamental Mechanism of Spring Fatigue Failure and Engineering Life Prediction Method
Introduction
The spring bears cyclic loads during service, and fatigue fracture is the most important failure mode, accounting for more than 80%. A seemingly intact spring can suddenly break after millions of cycles, often without any obvious signs before breaking. This "unwarned failure" is especially dangerous in safety-critical parts such as automobile suspensions, engine valves, and braking systems. For example, once the valve spring breaks, the valve will fall into the cylinder, causing the piston to penetrate the cylinder head and the engine to be scrapped instantly. If the suspension spring breaks, the wheel attitude can be out of control, and in severe cases, traffic accidents can be caused.
In 2025, a domestic OEM recalled a batch of new energy models due to multiple corrosion fatigue fractures in the rear suspension spring within 30,000 kilometers. Fracture analysis shows that the loss of control of shot peening process parameters leads to insufficient surface compressive stress depth, and the corrosive environment of winter snow melt salt reduces the spring life from the design target of 300,000 kilometers to less than 50,000 kilometers. This case reveals a key fact: the fatigue life of springs is not "measured", but "designed and manufactured".
Starting from the TCE-metal theory of fatigue failure, this paper systematically expounds the common types of spring fracture, fracture characteristics, and core factors affecting fatigue life, and gives practical life prediction methods and improvement measures for engineering.
The physical nature of spring fatigue failure
1.1 Three-stage evolution of fatigue cracks
Spring fatigue failure follows the classic three-stage model of "crack initiation, crack propagation, and instantaneous fracture":
Crack initiation stage (70% to 90% of the total life): Under repeated stress, microcracks form at the surface or subsurface of the spring (non-metallic inclusions, scratches, decarburization layer, shot peening indentation bottom, etc.). For high-stress springs, the initiation stage occupies most of the life.
Crack propagation stage (accounting for 10% to 30% of the total life): The crack propagates steadily at a rate of several microns per cycle, leaving typical fatigue glow patterns on the fracture surface (each glow pattern corresponds to one load cycle).
Instantaneous fracture stage (extremely short): When the remaining cross-section cannot withstand the peak load, the spring breaks rapidly with toughness or brittleness, forming a rough instantaneous fracture zone.
1.2 Key parameters affecting the fatigue limit (Basquin equation)
The fatigue life of a spring is usually described by a stress-life (S-N) curve. The Basquin equation gives a mathematical expression of the high-cycle fatigue region:
σ_a = σ_f‘ (2N_f)^b
Of which:
Sigma _a - stress amplitude
Sigma _f '- Fatigue strength coefficient (about 0.9 times tensile strength)
N_f - number of failed cycles
B - Fatigue strength index (typically -0.05 to -0.12)
Engineering experience shows that the fatigue limit of a spring is about 35% to 45% of its tensile strength, but this ratio will change significantly due to factors such as surface state, stress concentration coefficient, average stress, and environmental medium.
Second, the typical type of spring fracture and fracture identification
2.1 High-cycle fatigue fracture (most common)
Features: The fracture is flat, with clear fatigue source areas (often located on the inner surface of the spring), extended areas (smooth, with shell lines), and transient areas (rough, fibrous).
Reason: The design stress exceeds the fatigue limit of the material, or there is a source of stress concentration on the surface (e.g. indentation, scratches, decarbonization).
Typical case: The engine valve spring breaks after 10 ^ 8 cycles, and the source of fatigue is located at the rolling defect on the surface of the steel wire.
2.2 Corrosion fatigue fracture
Features: The surface of the fracture is covered with corrosion products (reddish-brown rust or black oxide scale), and the fatigue pattern is damaged by corrosion, and multi-source cracks often appear.
Reason: Under the combined action of corrosive medium (salt water, acid mist, electrolyte) and alternating stress, the fatigue limit decreases sharply or even disappears. Chloride solution can reduce the fatigue limit of springs by more than 50%.
Engineering countermeasures: switch to stainless steel or add coatings (Zn-Al dacromet, epoxy resin).
2.3 High temperature fatigue (creep-fatigue interaction)
Features: The fracture is accompanied by intergranular cracking and voids, and oxides can be seen at the grain boundaries.
Reason: In high temperature environment (> 500C) such as exhaust valve spring and turbocharger, creep and fatigue coupling accelerate failure.
Countermeasures for material selection: use nickel-based alloy (Inconel 718) or precipitation hardening stainless steel (17-7PH).
2.4 Hydrogen embrittlement fracture
Features: The fracture is characterized by intergranular brittle fracture, no fatigue glow pattern, and the crack propagates from the inside to the outside.
Reason: Hydrogen atoms infiltrated during pickling or electroplating accumulate under stress, causing the material to become brittle.
Mitigation measures: dehydrogenation baking within 4 hours after electroplating (200 C, ≥8 hours); use mechanical galvanizing or no hydrogen embrittlement coating (Dacromet).
Failure Type Fracture Characteristics Typical Environmental Life Reduction Proportion
High cycle fatigue single source, shell line, smooth extension zone drying, room temperature design life of 30%~ 50%
Corrosion fatigue Multi-source, rust, no clear glow pattern Salt spray, moisture, electrolyte design life 10% to 20%
High temperature fatigue intergranular cracking, oxide layer > 400C, 5% to 15% of the design life of gas environment
Hydrogen embrittlement intercrystalline brittleness, fatigue-free zone can break within a few hours without dehydrogenation after pickling/electroplating
III. Core engineering factors affecting spring fatigue life
3.1 Surface integrity (the most important factor)
More than 70% of spring fatigue sources are located at or near the surface. Therefore, surface integrity control is the primary means of improving lifespan:
Decarburization layer: The surface decarburization layer (ferrite) formed during heat treatment has extremely low strength and must be removed by grinding or shot peening. Allowable depth ≤ 0.05 mm.
Surface defects: scratches, indentations, folds, etc. produced during the spring winding process are equivalent to introducing sharp notches, and the stress concentration coefficient K_t up to 3 to 5.
Residual compressive stress: The residual compressive stress introduced by shot peening is "active protection". Experiments show that for every 100 MPa increase in surface compressive stress, the fatigue limit can be increased by about 30-50 MPa.
3.2 Stress concentration geometric characteristics
The shape of the spring itself has stress concentration: the inner stress is 1.2 to 1.6 times the average stress (depending on the winding ratio C = D/d). In addition, the end grinding, the transition area of the support ring, and the variable diameter are all stress concentration sensitive areas. Optimization suggestions: the winding ratio should not be less than 4; the transition angle between the support ring and the effective ring ≥ 0.5d.
3.3 Inclusions and cleanliness
Non-metallic inclusions (oxides, sulfides, silicate) in steel are potential sources of internal fatigue. For high stress springs, vacuum degassed steel or ESR steel is recommended with an inclusion grade of ASTM E45 ≤ 1.5.
3.4 Superposition of average stress and residual stress
According to Goodman's modified formula, the allowable stress amplitude _a decrease when the average stress sigma _m increases. The residual compressive stress sigma _r introduced by shot peening can be regarded as negative average stress, thus significantly increasing the allowable stress amplitude:
s _ a = s _ {-1} [1 - (s _ m + s _ r) / s _ b]
_ {-1} is the fatigue limit under a perfectly symmetrical cycle. When the residual compressive stress reaches -800 MPa, the effect is equivalent to canceling the average stress by 60% to 80%.
IV. Engineering practical life prediction method
4.1 Finite element simulation based on local strain method
Using elastoplastic finite element analysis, the stress-strain history of the spring danger point is calculated, and the crack initiation life is predicted by combining the strain-life (ε-N) curve of the material. Mainstream software includes ANSYS nCode DesignLife, FE-Safe, etc. Input parameters include:
Cyclic stress-strain curves of measured materials;
Surface roughness correction coefficient (generally 0.8~ 0.95);
Shot peening residual stress field (can be measured by X-ray diffraction and then loaded).
4.2 Fatigue test acceleration method
In order to shorten the test period, the lifting method or single-point method is often used to quickly evaluate the fatigue limit in engineering.
Lifting method: under the specified cycle base (e.g. 10 ^ 7 times), the stress level is changed step by step to obtain the median fatigue limit statistically.
Single-point method: Take 3 to 5 springs and test them under stress slightly higher than the estimated fatigue limit. If they all pass the base, the stress will be increased, and vice versa, the efficiency will be higher.
4.3 Actual life improvement case
A torsion bar spring for an automobile stabilizer has an original design life of 10 ^ 5 times (maximum stress 1,100 MPa). After the following measures are adopted, the life is increased to 210 ^ 6 times:
The material was upgraded from 60Si2MnA to 55CrSi (tensile strength increased from 1,800 MPa to 2,100 MPa).
Increase the stress shot peening once (increase the compressive stress from -400 MPa to -850 MPa).
The surface is coated with epoxy resin to prevent corrosion.
Corresponding to the increase in lifespan: 20 times.
V. Engineering proposals and checklists
5.1 Design Phase
Determine the target lifespan (number of cycles) and safety factor (generally 1.2 to 1.5);
Select the appropriate material grade and specify the grade of inclusions.
The stress distribution is analyzed by FEA, and the winding ratio and transition fillet are optimized.
Reserve a shot peening margin (0.1 to 0.2 mm diameter tolerance relaxation).
5.2 The manufacturing stage
Monitor the atmosphere of the heat treatment furnace and control the depth of the decarburization layer ≤ 0.05 mm;
Shot peening process verification: Almen strength, coverage, residual stress sampling test (XRD);
Do not bake after pickling or plating (risk of hydrogen embrittlement).
5.3 Acceptance and testing
Each batch of samples is taken for fatigue verification (at least 3 pieces).
For the use of spring in corrosive environment, add salt spray pre-corrosion + fatigue composite test.
conclusion
The fatigue failure of a spring is the result of a multi-factor coupling of material, manufacturing, design, and environment. Understanding fracture characteristics, controlling surface integrity, and choosing materials and strengthening processes rationally can increase the actual life of a spring from "well below the design value" to "beyond the design margin". For engineers, mastering S-N curves, residual stress theory, and failure analysis methods are essential skills to ensure the reliability of springs. The parameters, cases, and checklists given in this paper can be directly applied to daily engineering decisions.
BQUQ is a professional metal spring manufacturer, please send us drawings, and our company will quote you within 12 hours.
