What a Fatigue Fracture Looks Like: Difference from Static Fracture

Take a fatigued bolt for fracture-surface analysis and you almost always see three zones: fatigue origin (usually at the thread root, thread runout, or under-head radius—stress-concentration spots), fatigue propagation zone (smooth, with shell/clam-shell or beach markings), and the final instantaneous fracture zone (rough, smaller area). A static overload fracture, by contrast, has a rough full surface with obvious plastic necking. The field rule of thumb is simple: a flat smooth surface with clear clam-shell markings and a fracture zone taking only a small portion—basically fatigue. In engineering statistics, most bolt fatigue cracks start at the thread root, especially the first engaged thread with the nut (the most heavily loaded) and the thread runout. The beach or clam-shell ridges record the pause-growth rhythm of the crack under load, and their spacing narrows as the crack slows near final size; by measuring these, a failure lab can even estimate whether the joint saw hundreds of thousands or tens of millions of cycles before failure.

S-N Curve and Stress Amplitude: Mean Stress Does Not Decide Life

Fatigue life is described by the S-N curve (Wöhler curve): the higher the stress level, the fewer cycles it survives; when stress drops below a threshold, the bolt survives infinitely—this threshold is the fatigue limit. For a preloaded bolted joint, the key is not how high the mean stress on the bolt is, but the stress amplitude Δσ—the swing by which alternating load makes bolt tension fluctuate around the mean clamp force. Because the bolt is always preloaded near high tensile stress, external-load fluctuation merely superposes a small alternating component; it is precisely the size of this alternating component that drives fatigue-crack propagation.

Where the Stress Amplitude Comes From: Clamped-Part Stiffness and Load Distribution

The external force ΔF is not all carried by the bolt; it is "shared" by the stiffness of the bolt and the clamped parts. The alternating load actually taken by the bolt is ΔF_bolt = Φ·ΔF, where Φ is the load distribution factor (the bolt's share of stiffness): the more compliant the bolt and the stiffer the clamped parts, the smaller Φ, the smaller the stress amplitude the bolt gets, and the longer the life; conversely, a long thin soft bolt clamped in thick hard steel plates is comfortable, while a short stiff bolt clamped in thin soft parts is easily "forced" to carry the full alternating load. This is why engineers on vibrating equipment often deliberately use long thin bolts and add washers to enlarge the clamped-part bearing area—in essence all of them are tuning Φ smaller.

The Role of Preload: Why Tightening Hard Makes It More Durable

A counter-intuitive but key fact: sufficient preload is precisely the protection against fatigue. When the joint is fully pressed and external load is insufficient to separate the mating faces, the bolt's actual tension fluctuation is "buffered inside" the clamped parts, and ΔF_bolt stays small; once preload is insufficient and the mating faces repeatedly separate-and-reattach under alternating load, the bolt is forced to carry nearly the full alternating amplitude of external load, stress amplitude spikes, and fatigue life plummets. In addition, insufficient preload also causes slip wear and thread fretting, further accelerating fatigue. Of course preload is not "the bigger the better"—it must stay within yield and be paired with torque-clamp-force control to ensure it is built truly and stably.

Anti-Fatigue Design: Thread Rolling, Radii and Engagement Length

• Thread forming method: rolled threads form a continuous flow-line fiber during rolling and introduce residual compressive stress at the root, giving markedly higher fatigue strength than cut threads; fatigue-critical bolts always use rolled threads, preferably rolled after heat treatment.
• Thread root and radii: reduce root stress concentration—larger under-head radii and rounded runout threads (reducing runout stress concentration) are proven life-extension measures.
• Engagement length and nut matching: make thread failure occur at bolt tensile fracture rather than nut-end stripping; reasonable engagement length avoids overloading the first thread.
• Surface strengthening: shot peening and rolling introduce residual compressive stress at the root, further raising the fatigue limit.

Field experience reinforces the same lesson: when a bolted joint on vibrating equipment (pumps, gearboxes, vehicle chassis) fails in service, the root cause is rarely "weak material" but usually one of three—insufficient preload allowing face separation, a short stiff grip forcing the bolt to absorb the full alternating load, or a cut (machined) thread instead of a rolled one. Fixing any single lever often restores life; fixing all three at once is what VDI 2230 class-1 joints target. Conversely, simply buying a higher strength class—say upgrading from 8.8 to 12.9—does not move the fatigue limit much, because the fatigue limit of steel rises far more slowly than its tensile strength, and the extra hardness brings hydrogen-embrittlement risk.