Some observations on mode III fatigue thresholds

Abstract

The purpose of this note is to report some preliminary results from a study of near-threshold fatigue crack growth in a rail steel under mode III loading and, on the basis of these results, to make some general observations concerning the nature of mode III fatigue threshold. Table 1 contains the chemical composition and mechanical properties of the standard carbon rail steel used in this investigation.Edge-notched bar specimens having square cross-section were machined from the head of a heavy haul rail.Mode III fatigue tests were then conducted on these using a novel test rig developed at the University of New South Wales [i], which allows any desired combination of mode II and mode III loading conditions to be applied to the test specimen.The mode III fatigue threshold value was determined by an increasing load technique.Each specimen was tested for 4 million cycles at a given load range, with load ratio R_>0. If no evidence of fatigue cracking was found, either by the use of a specimen protection device sensitive to changes in the compliance of the specimen or by visual inspection of the notch with a monocular eyepiece, the specimen was re-tested at a higher load range.The results obtained by following this procedure are shown in Fig. I, from which a mode III fatigue threshold value for standard carbon rail steel of AKIIIt h = 18.8 ± 2.0 MPa/m is indicated.It is well established from torsion tests on cylindrical, circumferentially-notched specimens [2] that fatigue crack propagation under mode III loading generally leads to a macroscopically flat (mode III shear) fracture surface for high stress intensity ranges and short crack lengths, while a "factory roof" morphology (resulting from the growth of mode I branch cracks) develops at low stress intensity ranges.For a given applied loading, the growth rates of both types of crack diminish rather than increase with increasing crack length due to interference between the crack faces, and a transition commonly occurs from a macroscopically flat to a factory roof type of fracture surface. This transition has been taken to define a pseudo mode III threshold for macroscopically flat fatigue crack growth [3], the value of which is in general considerably higher than the fatigue threshold corresponding to factory roof cracking. Pook and Sharples [4] have suggested from consideration of a mode Ibranch at a mode III crack that the ratio of this latter mode III threshold to the mode I threshold for a given material should theoretically be 1.35. The value of AK . reported in the present note is larger by a factor of about 1.4 than ~I t~ " Table I, in e mean value of AK h for rail steel glven in t fairly good agreement with this pre~iction. 29 (1985) R45 R46 Examination of the fatigue fracture surfaces by scanning electron microscopy revealed the presence of a factory roof morphology in all cases, as expected from the considerations outlined above. However, the fracture surfaces of some specimens also exhibited a number of small, macroscopically flat regions where fatigu...

Mixed Mode (I + Iii) Fatigue Thresholds in a Forging Steel

“…Pook [9] noted that different thresholds could be found for microscopic crack growth, crack arrest and specimen failure. Hellier et al [12] similarly defined the mode I11 threshold as the boundary between no crack growth and facetted crack growth. Evaluation of branch crack stress intensity factors is difficult.…”

“…Substituting equations (8) and (10) into equation(12) and considering the cyclic values of the stress intensity factor range, thenA No facets Facetsz] C O S~ +2 25. [ -i:::] sin 4 = 1…”

The Effect of Mean Stress on Mixed Mode (I + Iii) Fatigue Thresholds

“…Fractographic studies of mixed mode (I + 111) fatigue specimens have shown that fine mode I facets are formed close to the slit tip and in some cases [5,7] co-planar crack growth at the crack tip occurs, similar to the shear cracks observed in mode 111 [9,10] and in mixed mode (If 11) fatigue [I 11.…”

“…Unlike mode I fatigue testing there is no generally accepted definition of the fatigue threshold under mode I11 or mixed mode conditions. Thresholds have been variously defined by the lowest stress intensity factor range at which mode I11 flat crack growth occurred [12], or the lowest stress intensity factor range for the formation of mode I branch cracks or facets [2,7,10,13,14]. Different thresholds can be found for microscopic crack growth, crack arrest and specimen failure.…”

Estimation of Torsional Fatigue Strength of Medium Carbon Steel Bars With a Circumferential Crack by the Cyclic Resistance‐curve Method

“…As for the threshold condition for the branch (mode I) crack formation, Pook [5] proposed the maximum principal criterion, while Yates and Miller [6] proposed the critical crack-tip opening displacement (mode I) criterion. Yates and Miller [6], and Hellier et al [7] observed small microscopic flat fracture surfaces very near to the initial notches. The condition for the formation of this flat fracture is not yet known.…”