Material characterisation and modelling of fatigue initiation in aluminium based plain bearing materials

Abstract

This thesis first presents the results of a three-year research programme characterising the elasto-plastic material properties of the thin A1-Sn based alloy linings used in modern plain journal bearings applications.  Then it presents work on the investigation of the effect of microstructure (particularly particle distributions) on fatigue initiation behaviour in linings.

For material characterisation of thin lining layers for which the stress strain data cannot be obtained through traditional tensile tests, a methodology was proposed involving a combination of different experimental and numerical techniques: Vickers micro-hardness test, FEM, adaptive numerical modelling and genetic algorithms.  In this methodology a stress-strain curve that could give an identical load-indentation curve from FE modelling as the experimentally obtained load-indentation curve from Vickers micro-hardness tests was taken to be a good approximation of the property of that material.  Adaptive numerical modelling and genetic algorithms were used to perform the iterations and optimisations to identify such a stress-strain curve.  The generic nature of the procedure of this methodology allows it to be extended to other materials, rather than just A1 linings.

FE models of the micro-structure and classification approaches using adaptive numerical modelling techniques were applied to investigate the effect of critical microstructure parameters, i.e. particle distributions, on initiation behaviour in A1 linings. In the FE modelling (using materials models obtained from the material characterisation procedure detailed earlier) the maximum plastic shear strain and hydrostatic stress were adopted as possible initiation criteria.  Microstructure features revealed by an in-house tessellation program on fatigued linings were compared with FEM results.  FE modelling of the real microstructure shows correlation between crack initiation sites and the high local stress/strain initiation criteria.  The classification models on the preferential crack initiation particles observed experimentally provided possible predictions of microstructural sites most likely to initiate fatigue.

More information

Identifiers

Catalogue record

Export record

ASCII CitationAtomBibTeXData Cite XMLDublin CoreDublin CoreEP3 XMLEndNoteHTML CitationHTML CitationHTML ListJSONMETSMODSMPEG-21 DIDLOpenURL ContextObjectOpenURL ContextObject in SpanRDF+N-TriplesRDF+N3RDF+XMLRIOXX2 XMLReferReference ManagerSimple Metadata