| Summary: |
Adhesive joints have been intensively investigated over the past 60 years. The first paper on stress analysis of adhesive joints was by Volkersen in 1938. In the shear lag analysis of Volkersen, the adhesive was subjected to shear loading caused by the differential straining of the adherends. But there was an important point that was neglected in the Volkersen analysis: the rotation of the joint. This point was included in the Goland and Reissner analysis in 1944. Several improvements were made over the second part of the 20th century by Hart-Smith (1973) and Allman (1977) for example.
The first joint strength predictions were based on elastic stress analysis and it was found that the joint strength prediction was well below the experimental results. In 1973, Hart-Smith modelled the adhesive taking into account its plastic behaviour and obtained more realistic joint strength predictions.
To have successful joint strength predictions, the model must take into account the geometrical and material non-linearities (Harris and Adams, 1984). When bonding composites, special care must be taken since the mechanics of failure are more complex due to the low through thickness strength of the composite (Adams and Davies, 1996).
The close-form solutions described above, whether simple or complex, introduce a certain degree of simplification. The finite element method is a more complete tool for stress analysis since all the stresses can be easily calculated without the need to introduce restrictive assumptions as in a closed-form analysis (Adams et al., 1997). However, there are some disadvantages too, such as the problem of stress singularity at sharp corners (da Silva and Adams, 2007). The finite element method is used as a research tool, when for example a complicated joint geometry is involved, and the closed-form analysis is more relevant in terms of practical design.
There are many closed-form solutions available to the designer. However, there are important practi  |
Summary
Adhesive joints have been intensively investigated over the past 60 years. The first paper on stress analysis of adhesive joints was by Volkersen in 1938. In the shear lag analysis of Volkersen, the adhesive was subjected to shear loading caused by the differential straining of the adherends. But there was an important point that was neglected in the Volkersen analysis: the rotation of the joint. This point was included in the Goland and Reissner analysis in 1944. Several improvements were made over the second part of the 20th century by Hart-Smith (1973) and Allman (1977) for example.
The first joint strength predictions were based on elastic stress analysis and it was found that the joint strength prediction was well below the experimental results. In 1973, Hart-Smith modelled the adhesive taking into account its plastic behaviour and obtained more realistic joint strength predictions.
To have successful joint strength predictions, the model must take into account the geometrical and material non-linearities (Harris and Adams, 1984). When bonding composites, special care must be taken since the mechanics of failure are more complex due to the low through thickness strength of the composite (Adams and Davies, 1996).
The close-form solutions described above, whether simple or complex, introduce a certain degree of simplification. The finite element method is a more complete tool for stress analysis since all the stresses can be easily calculated without the need to introduce restrictive assumptions as in a closed-form analysis (Adams et al., 1997). However, there are some disadvantages too, such as the problem of stress singularity at sharp corners (da Silva and Adams, 2007). The finite element method is used as a research tool, when for example a complicated joint geometry is involved, and the closed-form analysis is more relevant in terms of practical design.
There are many closed-form solutions available to the designer. However, there are important practical cases that do not have a closed-form solution. These include joints with spew fillets, joints with adherend shaping and joints with two or more adhesives in the overlap. During manufacture, there is always part the adhesive squeezed out of the joint forming a spew fillet, and it has been shown that it has a great influence in the stress distribution (Crocombe and Adams, 1981). Another important aspect is adherends tapering. Adams et al. (1986) and da Silva and Adams (2007) have proved that it is a very powerful technique for reducing the peel stresses in composite adherends. Finally, the use of a joint with two adhesives can be a useful means of increasing the joint strength of a joint with a brittle adhesive (Pires et al., 2003), or the way to have a joint with a uniform strength from low to high temperatures (da Silva and Adams, 2007).
Adhesive joinning is a method of joining in expansion but is still mainly used by high technology industries (aerospace for example). The main reason for the limited use of this promising technique is the lack of realistic design tools. The literature contains many closed-form solutions for the stress distribution. However, these solutions are for regular geometries, with only one adhesive in the overlap and no spew fillet. These aspects have a very important influence in the joint mechanics and the main objective of this project is to develop a closed-form solution that incorporates them. The closed-form solution will be made available in the form of a user friendly software, readily available to the design engineer. The analytical model will be validated with a finite element analysis and experimental tests. |