A.M. Pereira

Influence of Superposition Length on Transverse Impact Response of Single-Strap Adhesive Joints

Adhesive joints are usually designed to carry in-plane loads, but in many cases they are also prone to transverse loading. On the other hand, the impact response of adhesive joints has received limited attention compared to quasi-static loading. Therefore, the present paper aims to study the influence of superposition length on transverse impact response of single-strap adhesive joints. For this purpose, low-velocity impact tests were performed using a drop weight-testing machine with a hemispherical impactor falling at the center of a bi-clamped specimen. The specimens were manufactured using Docol 1000 high-elastic limit steel, with 1.5 mm of thickness, and an Araldite® 420 A/B adhesive (Huntsman Advanced Materials, Everberg, Belgium). The collapse thresholds obtained were 61.6 J, 75.1 J, and 77.5 J, respectively, for adhesive joints with gap length of ℓo = 0, 10, and 20 mm. An adhesive fracture occurred for the three geometries and the cracks initiated at the corner of the joint where the deflection is higher. Joints with higher ℓo have higher impact energies, despite the lower bonding area, as consequence of the lower local deformation. A numerical study was developed and the zero gap (ℓo = 0) gives maximum peel stress.

Assessment of the fatigue life of aluminium spot-welded and weld-bonded joints

In modern machinery and automobile structures weight reduction and increased durability are the main issues in design. In these applications, lap welded and/or bonded joints are widely used; therefore, tools are needed to accurately predict their fatigue life. This paper is concerned with the fatigue strength of single lap joints formed with thin plates of 6082-T6 aluminium alloy using a high strength two-component epoxy adhesive (Araldite 420 A/B from Hunstman). Experimental S–N curves were obtained for resistance spot-welded and weld-bonded lap joints. The fatigue lives of weld-bonded joints were significantly higher than those of resistance spot-welding joints. In addition, fatigue lives were predicted with Morrows modified Manson–Coffin (M/M–C) and the Smith–Watson–Topper (S–W–T) damage equations. Elastic–plastic numerical models were developed, replicating the experimental work, in order to obtain local stress and strain fields. An acceptable agreement was obtained between the numerical predictions and the experimental results. The M/M–C damage equation diverged from experimental results for relatively long fatigue lives, while the S–W–T equation gave good predictions for all fatigue lives.

Effect of the mean stress on the fatigue behaviour of single lap joints

ABSTRACT Steel is the most important construction material for the mass production of engineered structures, especially in the transport industry. On the other hand, adhesive joints are typically used to join load-bearing components. Therefore, this work intends to investigate the stress ratio effects on the fatigue behaviour of adhesively bonded steel lap joints. S–N diagrams of fatigue tests, under constant amplitude loading, were obtained for stress ratios ranging between 0.05 and 0.7. It was observed that the fatigue life of the adhesive joints has very little dependence on the stress amplitude, indicating that only the maximum stress is important. The combination of a linear equation with a quadratic equation seems to be the best formulation to fit the experimental results. Finally, the Palmgren–Miner’s Law is accurate enough to predict the fatigue design for sequential block loadings.

Cyclic creep response of adhesively bonded steel lap joints

ABSTRACT The viscoelastic nature of polymeric adhesives means that the effect of fatigue frequency has to be treated cautiously. However, this subject has received limited attention and very few studies can be found. Therefore, this work aims at investigating the cyclic creep response of adhesively bonded steel lap joints. Load-controlled fatigue tests were performed with shear stresses of 9.1, 7.4, and 6.3 MPa, which are typically low cycle fatigue stresses. Only during the last 20% of fatigue life can we observe an increase in the cycle hysteresis area due to the decrease of the shear stiffness caused by the failure mechanisms. Under fatigue load, the maximum/minimum strain curves exhibit a shape being similar to that of the steady creep curves, in which occurs a second stage with nearly constant strain rate, independently of the number of cycles and increasing with the load range. A linear relationship between the log cyclic creep rate and the log of the number of cycles to failure was observed, indicating that fatigue behaviour is strictly related to cyclic creep.