Fatigue failure — crack initiation and propagation under cyclic loading below the static yield strength — is the dominant failure mode for structural joints in vehicles, aircraft, machinery, and infrastructure. A joint that carries its design load statically with a factor of 3 safety margin can still fail in fatigue after millions of cycles if the stress concentrations within it are high enough. The reason bonded structural joints consistently outperform mechanically fastened joints in fatigue is not adhesive chemistry — it is load path geometry. Understanding this advantage, and the conditions that can compromise it, is essential for specifying structural epoxy in fatigue-loaded applications.
Why Bolted Joints Fail in Fatigue
The Achilles heel of bolted and riveted joints in fatigue is the stress concentration at the fastener hole. At a circular hole in a plate under uniaxial tension, the peak tangential stress at the hole edge is three times the nominal plate stress — a stress concentration factor Kt of 3. Under cyclic loading, the fatigue crack initiates at the peak stress location — the hole edge — and propagates through the net section.
The practical consequence: bolted joints in aluminium structure typically fail in fatigue at nominal stresses well below the material’s fatigue endurance limit. The stress at the hole edge exceeds the local fatigue threshold even when the nominal stress is considered safe. This is why aircraft maintenance programs require extensive fastener hole inspection — fretting under the fastener head, combined with the stress concentration, creates a reliable fatigue crack initiation site.
Increasing plate thickness to reduce nominal stress reduces the stress amplitude at the hole proportionally, but the stress concentration factor remains 3. Fatigue life improvements from thickness increase in bolted joints are thus less efficient than the proportional stress reduction suggests.
Why Bonded Joints Perform Better in Fatigue
Structural epoxy bonds transfer load through shear distributed over the full bond area. There is no hole, no fretting contact surface, and no discrete stress concentration. The shear stress is highest at the overlap ends due to elastic shear lag, but the stress distribution — while non-uniform — is a smooth gradient rather than a factor-of-3 stress concentration at a point.
For the same nominal applied stress, the peak stress in a well-designed bonded joint is lower than in an equivalent bolted joint. Combined with the absence of fretting (which accelerates fatigue crack initiation at fastener contacts), bonded joints consistently show longer fatigue life in controlled comparisons.
Published fatigue test results for bonded vs. riveted aluminium lap joints at equivalent bond/fastener area show:
– At high stress levels (>60% of static strength): bonded and riveted joints have similar fatigue life
– At moderate stress levels (30–50% of static strength): bonded joints survive 3 to 10 times more cycles
– At low stress levels (10–20% of static strength): bonded joints approach a fatigue endurance limit; riveted joints continue to accumulate damage
The endurance limit behavior at low stress is particularly valuable in applications where the number of cycles is very large (rotating machinery, aircraft pressurization cycles) — bonded joints can survive indefinitely at stresses below a threshold; bolted joints do not show this behavior because fretting prevents a true endurance limit.
If you need fatigue S-N data for structural epoxy on specific substrates, comparative bolted-vs-bonded fatigue test results, and adhesive recommendations for cyclic loading applications, Email Us — Incure provides fatigue characterization data for structural adhesive qualification.
Factors That Degrade Bonded Joint Fatigue Performance
Bonded joints in fatigue are not unconditionally superior to fastened joints — there are conditions where the advantage is reduced or eliminated:
Peel at bond edges. If the joint design generates peel stress at the bond edge — from eccentric loading, thermal cycling, or vibration out-of-plane modes — fatigue crack initiation occurs at the bond edge where the peel stress is concentrated. A bonded joint with significant peel stress may show fatigue life comparable to a bolted joint because the bond edge peel stress concentration replaces the fastener hole concentration as the fatigue initiation site.
Bond line defects. Voids, areas of poor adhesion, or disbonds within the bond line act as pre-existing flaws that concentrate stress in the surrounding adhesive. Fatigue crack propagation from a void in the bond is more rapid than from a clean initiation site because the void is an existing stress concentrator within the adhesive. Void-free bonding is a requirement for fatigue-critical applications.
Moisture degradation at the bond interface. Long-term humidity exposure degrades the adhesive-substrate interface through moisture displacement of the adhesive from the metal oxide surface. As the effective bond area decreases over time due to interfacial disbondment from the bond edge, the remaining bond carries higher stress — accelerating fatigue damage. The fatigue life of a well-prepared bond in dry service is not representative of the same bond after 10 years of outdoor humidity exposure.
Design Approach for Fatigue-Critical Bonded Joints
Taper the overlap ends. Tapering the adherend at the bond end reduces the shear stress concentration and reduces peel stress. This is the highest-leverage geometric change for improving bonded joint fatigue life.
Use toughened adhesive. Toughened epoxy with rubber particles or core-shell particles resists fatigue crack propagation through the adhesive by crack tip blunting and energy absorption. Under cyclic peel or mode-I loading, toughened adhesive shows dramatically better fatigue crack propagation resistance than unfilled epoxy.
Qualify with fatigue testing, not static lap shear alone. Static lap shear strength is a poor predictor of fatigue life — two adhesives with identical static strength can have very different fatigue performance depending on modulus, toughness, and failure mode under cyclic loading. For applications where fatigue governs, fatigue S-N testing on the actual joint geometry and substrate is required for design qualification.
Contact Our Team to discuss toughened adhesive selection, joint geometry optimization for fatigue, and fatigue qualification test planning for structural epoxy applications under cyclic loading.
Visit www.incurelab.com for more information.