Search results

1 – 1 of 1
Article
Publication date: 4 January 2008

Dimitri Leray, Aurelian Vadean and Alain Daidie

This paper seeks to deal with a new modelization method which aims at fatigue sizing of preloaded bolted joints. Industrial design offices indeed need new models which, on the one…

Abstract

Purpose

This paper seeks to deal with a new modelization method which aims at fatigue sizing of preloaded bolted joints. Industrial design offices indeed need new models which, on the one hand, take bending of the bolts and geometrical non‐linearity into account and, on the other hand, run fast enough to be used for preliminary design stages. Usual sizing procedures derive from VDI recommendations, which makes them inaccurate. On the contrary, classical finite element methods are revealed to be very costly.

Design/methodology/approach

The first task lies in reducing the physical problem down and model the structure using axisymmetrical elements. Then, the core of the method lies in modifying the stiffness matrix of a tube element, in order to modify the axial compression stiffness to the one used by preloaded assembly classical computations. Eventually, a 2D finite element model is programmed which takes advantage of the modified element. A mounting was built to reproduce the typical loading of a slewing bearing. Experimental tests were carried out in order to help analyse the problem and to check finite element simulation results.

Findings

Sample experimental results are presented which confirm the need for new models and validate the 2D model that was developed.

Research limitations/implications

The new finite element, as well as the set of hypotheses that are used, appear to be usable for other bolted joints.

Practical implications

A software was produced for the industrial partners, which is usable by non FE‐specialists.

Originality/value

This work may serve as a basis for building fast and accurate finite element models of other types of bolted joints.

Details

Engineering Computations, vol. 25 no. 1
Type: Research Article
ISSN: 0264-4401

Keywords

1 – 1 of 1