Modelling of combined roughness and plasticity induced closure effects in high strength Al-alloys

Abstract

An investigation of plasticity induced crack closure (PICC) and roughness induced crack closure (RIC C) behaviour using finite element (FE) methods is presented for cracks subjected to small scale yielding (SSY) conditions. For constant amplitude (CA) undeflected cracks have been examined under both plane strain and plane stress conditions, whilst plane strain analyses have been particularly considered for deflected cracks. A previous two dimensional analytical treatment of RICC (2D CA RICC) [Parry, 2000] has been extended to produce a 'continuous' closure model matching the FE findings. The model is further modified to address three dimensional effects and compared to detailed experimental findings. Results shows the increase in closure levels with increasing twist angle (rjJ) are less significant compared to that with increasing tilt angle (8). Further FE modelling of PICC and RICC for cracks subjected to single overloads is presented. A single overload analytical model of PICC proposed by Parry has been modified following similar arguments to the CA-RICC model. The analytical model has further been modified to address RICC effects during single overloads. Effects of Llrp ratios on deflected cracks during overloads are seen to be functionally similar to RICC under constant amplitude loading (particularly in the 'saturation' of RICC influence for Llrp(OL) > 1), where L, rp and rp(OL) are deflected crack length, baseline plastic zone size and overload plastic zone size respectively. Competitive influences of PICC and RICC effects during single overloads are identified in both the FE and simplified analytical models. A modified 'strip yield' analytical model of the 'FASTRAN'-type [Xu, 2001] has also been used to study PICe effects during single overloads. Comparisons of models and experimental closure and growth rate studies are presented. Investigations have then been extended to consider both double and multiple overload conditions. In particular, attempts are made to study the effects of overload spacing on closure levels and growth rates. It appears to exist a maximum overload interaction zone for double overloads which severity of overload closure effect is at its greatest. Overall it is found that key functional aspects of the various FE models are reproducible in simple analytical representations of RICC and PICC efforts. Whilst some fitting is involved, good correlation of the present analytical models and experimental data is shown, opening a potential route to improve, computationally efficient, multimechanistic fatigue lifing methods involving crack closure.

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