The effect of machined surface topography and integrity on fatigue life

Donka Novovic, Richard Dewes, David Aspinwall, W Voice, Paul Bowen

Research output: Contribution to journalArticle

275 Citations (Scopus)

Abstract

The paper reviews published data which address the effect of machining (conventional and non-conventional processes) and the resulting workpiece surface topography/integrity on fatigue performance, for a variety of workpiece materials. The effect of post-machining surface treatments, such as shot peening, are also detailed. The influence of amplitude height parameters (Ra, Rt), amplitude distribution (Rsk) and shape (Rku) parameters, as well as spatial (Std, Sal) and hybrid (Ssc) measures, are considered. There is some disagreement in the literature about the correlation between workpiece surface roughness and fatigue life. In most cases, it has been reported that lower roughness results in longer fatigue life, but that for roughness values in the range 2.5-5 mum Ra it is primarily dependent on workpiece residual stress and surface microstructure, rather than roughness. In the absence of residual stress, machined surface roughness in excess of 0.1 mum Ra has a strong influence on fatigue life. Temperatures above 400 degreesC reduce the effects of both residual stress and surface roughness on fatigue, due to stress relieving and the change in crack initiation from the surfaces to internal sites. The presence of inclusions an order of magnitude larger than the machined surface roughness generally overrides the effect of surface topography. (C) 2003 Elsevier Ltd. All rights reserved.
Original languageEnglish
Pages (from-to)125-134
Number of pages10
JournalInternational Journal of Machine Tools and Manufacture
Volume44
Issue number2-3
DOIs
Publication statusPublished - 2 Sept 2004

Keywords

  • surface integrity
  • surface topography
  • fatigue life

Fingerprint

Dive into the research topics of 'The effect of machined surface topography and integrity on fatigue life'. Together they form a unique fingerprint.

Cite this