You are being redirected because this document is part of your ASTM Compass® subscription.
    This document is part of your ASTM Compass® subscription.

    If you are an ASTM Compass Subscriber and this document is part of your subscription, you can access it for free at ASTM Compass

    Fatigue Prediction Based on Computational Fracture Mechanics

    Published: 01 January 1996

      Format Pages Price  
    PDF (272K) 16 $25   ADD TO CART
    Complete Source PDF (6.8M) 354 $131   ADD TO CART

    Cite this document

    X Add email address send
      .RIS For RefWorks, EndNote, ProCite, Reference Manager, Zoteo, and many others.   .DOCX For Microsoft Word


    A general numerical method has been developed to predict fatigue crack initiation and fatigue crack propagation under cyclic loads. Finite element technique is employed to model specimen geometry. The method requires uniaxial stress/strain material data only; empirical fatigue constants are not required. It determines the number of load cycles for crack initiation or for crack propagation based on fracture energy, W, and dissipated energy per cycle, E. Fracture energy is the plastic-strain energy contained in a Dugdale-type cohesive stress zone located near an initial defect. The size of the cohesive zone of a specimen is established by a computational procedure that simulates elastic-plastic fracture under static loading. Dissipated energy per cycle is the plastic-strain energy in a hysteresis loop that exhibits cyclically stable behavior after “shake-down.” The number of cycles to initiate a crack or to propagate a crack to a finite amount is obtained by dividing the fracture energy, W, by the dissipated energy per cycle, E.

    Verification examples showing stress amplitude versus number of cycles (S-N diagram) are compared with experimental results. Also, experimental crack growth rate as a function of stress intensity factor (Paris-type equation) is compared with calculated results based on this method. Fatigue constants based on experimental data were not required in the computations. Limited comparison of predicted cycles to experimental results indicates good correlation. This work is significant in that it suggests a new and improved approach to problems of fatigue. Since fatigue behavior can be predicted using only uniaxial material tensile data, it holds out the promise of a reduced need for experimental work in the future.


    fatigue prediction, fracture mechanics, testing

    Author Information:

    Chang, AT
    Stevens Institute of Technology, Hoboken, NJ

    Nelson, NW
    Stevens Institute of Technology, Hoboken, NJ

    Cordes, JA
    Stevens Institute of Technology, Hoboken, NJ

    Kim, Y-J
    Stevens Institute of Technology, Hoboken, NJ

    Committee/Subcommittee: E08.06

    DOI: 10.1520/STP16134S