STP1074

    Prediction of Fracture Toughness by Local Fracture Criterion

    Published: Jan 1990


      Format Pages Price  
    PDF (452K) 17 $25   ADD TO CART
    Complete Source PDF (13M) 17 $132   ADD TO CART


    Abstract

    Incorporation of micromechanistic criteria of failure to analytical or numerical crack tip stress and strain solutions, known as the RKR (Ritchie, Knott, and Rice) model, derives a certain relation between the fracture toughness and flow/fracture properties of materials. Based on the above analytical prediction, correlations between the fracture toughness and yield stress, cleavage fracture stress, and critical plastic strain of the materials have been investigated for cleavage and ductile fracture. Nine types of low carbon structural steel were tested that had various microstructures with yield strengths of 280 to 1110 MPa. Good correlations according to analytical predictions have been obtained for both cleavage and ductile fracture, and the possibility of quantitative prediction of the fracture toughness from a conventional round bar tensile test can be shown. Effects of metallurgical and mechanical factors on the fracture toughness, and temperature dependence of the fracture toughness, can be explained from their effects on the flow and fracture properties.

    Keywords:

    fracture toughness, local fracture criterion, RKR model, cleavage fracture, ductile fracture, low carbon steels, cleavage fracture stress, critical plastic strain


    Author Information:

    Miyata, T
    Associate Professor, Nagoya University, Nagoya,

    Otsuka, A
    Professor, Nagoya University, Nagoya,

    Mitsubayashi, M
    Graduate StudentAssociate Engineer, Toyota Motor Corporation, Toyota,

    Haze, T
    Senior Researcher, R&D Laboratories II, Nippon Steel Corporation, Sagamihara,

    Aihara, S
    Research Engineer, R&D Laboratories II, Nippon Steel Corporation, Sagamihara,


    Paper ID: STP19004S

    Committee/Subcommittee: E08.04

    DOI: 10.1520/STP19004S


    CrossRef ASTM International is a member of CrossRef.