Structural modelling of adhesive joints in automotive bodies

The behaviour of a vehicle is dependent on the stiffness of the joints between the sheet components in the body shell. Although spot-welding is the predominant joining method for the construction of steel bodies in the automotive industry, other alternatives are now being considered. Adhesive bonding offers the best potential because the uniform load distribution in the joint reduces the stress concentrations produced when using spot-welding. It is generally observed that adhesive bonded structures are stiffer than assemblies fabricated with mechanical fasteners or spot-welds. The use of adhesives, either as an alternative or a supplement to spot-welding, is of interest because of the potential improvements in joint stiffness and in the overall behaviour of the structure. The effective stiffness of an adhesively bonded joint may be difficult to quantify, as it is dependent on many design variables of the actual joint. Finite element models have been developed to study the effects of various design parameters, such as different joining techniques, on car body characteristics. However, finite element models of large vehicle structures involve large numbers of elements which as a result can impose excessive demands in computer capacity. Because of this, approximations are commonly made in the model to reduce the number of elements consequently resulting in inaccuracies. The main inaccuracy develops from the lack of geometric details within local joints which may lead to some uncertainty in local joint stiffness. As a consequence, this may introduce errors in the prediction of global vehicle stiffness. In this project, the errors which result from the inaccuracies in macro modelling methods have been addressed through a parallel study of detailed micro models. Various adhesive joint configurations have been analysed using finite element methods to predict joint stiffness; comparative results were obtained through experimental testing of similar joints. A primary objective was to obtain characteristics of smaller joint structures and compare these to results from macro models of similar joints which could then be translated into larger-scale structure models for improved accuracy. A methodology has been developed to translate micro model joint characteristics into large- scale structures through an undercut element technique. When applied to macro models which lack geometric details of joints, the undercut element method enables more accurate predictions of behaviour, particularly stiffness performance, to be made. The method has been shown to be applicable to a number of typical joints and also for different loading conditions. Validation of the method has been demonstrated by its application to progressively larger substructures, from which experimental test data was available for comparison. Because of the ease of use of the undercut method it may be conveniently applied to automotive bodies. The method provides a more accurate finite element model resulting in less computational time for analysis than other existing solutions.

Grantor dc:publisher

Oxford Brookes University

Year dc:date

2000

Author dc:creator

• Steidler, Silvana M.

Contributors dc:contributor

• Beevers, Alec
• Durodola, John

Statement dc:rights

• All rights reserved

Language dc:language

en

DOI dc:identifier

https://doi.org/10.24384/hzrf-yh39

OAI identifier oai:identifier

tle:976394e5-bb8e-489e-a46e-24c20e6c35e1:d6bd9758-527a-46cd-bfe2-c433766e8fca:1