Le comportement mécanique fin d’un assemblage collé

Analyse expérimentale et théorique dans le cas du double recouvrement biseauté^*

Responsable de recherche, laboratoire de Science des Matériaux, Université Sciences Exactes et Naturelles de Reims Champagne. Chef du Groupe de recherche : « Composites et Collage », laboratoire de Mécanique et Énergétique, École Nationale Supérieure de Techniques Avancées, Palaiseau. Maître de Conférences à l’Université de Reims Champagne-Ardennes (IUT de Reims/département génie civil).

Abstract

We compare the fine mechanical behaviour of two adhesively bonded joints, of the double-lap type, one of which has tapered splice plates.

The active surfaces (bearers) of the adherends underwent a chemical sulfochromic treatment, directly applied to the raw surfaces of 2024-PLA5 Aluminium alloy in T₃ state, in conformity with the recommandations of Aerospatiale Aquitaine.

The roughness geometry of bearers is defined by the principal parameters of profile, obtained through mechanical palpation, and by the Abbot curve.

Mechanical behaviour of the test specimens is experimentally characterized in the course of a trial in simple traction, by the simultaneous measurement of deformations through electrical strainganges, and the record of acoustical emission. The trials are realized at room temperature.

The studied structures have a very complex behaviour. This sterns from the fact that the adherends are not slender (overlap length l = 12,5 mm, overlap width l₁ = 25 mm), and that they are bonded by a supported, adhesive film, of Hysol 9628 NW, whose behaviour is strongly elastic-plastic.

This situation induced us to use, for the modelling, a slender assembly, fully bevelled, the adherends of which are bonded by a brittle adhesive without marked visco-elasticity. A simplified calculation has been worked out by Michel L. L. Klein, by utilizing a method that generalizes the classical theory, where the adherends are strained in preponderant traction (see pp. 33-36 of this issue).

If we take, for darnage criterion, the begining of the first microcracks, it becomes apparent that the test-specimen with tapered splice plates is the most resistant. The results, obtained by the two techniques, are very correlated.