Stress-intensity threshold conditions for no crack growth of small cracks at the impact site were defined based on the relationship of Lukáš [9] for small cracks at notches; for further details see Ref. [5]. Because the FOD-initiated microcracks have lengths (2c ~ 1 to 50 µm) that are comparable to the scale of local plasticity, there may be a degree of uncertainty in the calculation of their driving forces. However, maximum plastic zone sizes are only ~ 0.2 to 1 µm for 1 to 10 µm sized cracks at ∆K ~ 1 - 2 MPa√m, such that conditions are close to that of small-scale yielding. The contribution of the residual stresses is not included in this ∆K calculation since these stresses do not change the stress intensity range. They do, however, affect the mean stress and hence alter the local load ratio, as is examined below. Indeed, initial estimates of the magnitude and gradient of the residual stresses have been computed [7] and are experimentally measured in this study. RESULTS AND DISCUSSION Simulation of foreign object damage Damage sites on the flat surface of the rectangular gauge section of KB specimens, resulting from 300 m/s impacts with 1 and 3.2 mm steel shot, are shown in Figs. 1a,b. Increasingly severe damage states have been previously reported with increasing impact velocity for the bi-modal Ti-6Al-4V [4,5]. For impact velocities in excess of ~250 m/s, pile-up of material (Fig. 1) at the crater rim resulted in multiple formation of microcracking (insert in Fig. 2). Moreover, circumferentially-oriented intense shear bands were formed, emanating from the surface of the impact crater. The microcracks (~2 to 50 µm) were found to provide the nucleation sites for crack growth on subsequent cycling (Fig. 2), although such microcracking was not seen at slower 200 m/s impacts. For this reason, we believe that low velocity or quasi-static indentations do not provide a realistic simulation of FOD. In addition to the formation of the FOD-induced microcraks, the presence of tensile residual stresses at locations remote from the impact site was also of importance for crack initiation, especially for failures in the 107 to 108 cycles regime. These tensile stresses, measured using synchrotron x-ray micro-diffraction techniques at the side faces of the KB specimens, counterbalance the compressive residual stress field directly beneath the indentation, as predicted numerically [7]. The estimated distributions of these FOD-induced stresses in KB specimens are shown in Figs. 3a and b for 300 m/s impacts using 1 mm and 3.2 mm steel shot, respectively, and display tensile residual stresses (in longitudinal direction) as high as ~220 MPa ( σ33/ σy = 0.24) to 385 MPa ( σ33/ σy = 0.42), where σy is the yield stress, and σ33 the normal residual stress in longitudinal direction. The effect of such residual stresses on fatigue crack initiation and the fatigue crack growth thresholds is discussed below. Fatigue properties Stress-life (S-N) data in Fig. 4 clearly show the detrimental effect of FOD on the fatigue strength of bi-modal Ti-6Al-4V. The reduced lifetimes following impact damage was seen for both projectile sizes and at both low and high load ratios (R = 0.1 and 0.5). As discussed previously [8-10], this detrimental effect of FOD can be considered in terms of premature fatigue crack initiation resulting from (i) the stress concentration of the indent, (ii) FOD-induced microcracking, (iii) plastically-deformed microstructure and (iv) tensile residual stresses that result from the impact. Two groups of failures can be identified in Fig. 4, dependent upon the magnitude of the applied stresses relative to the tensile residual stresses. At high applied stresses, HCF failures initiate directly at the impact crater. For 300 m/s impact velocities where FOD-induced microcracks were formed, fatigue cracking initiated from these microcracks at the crater rim (surface crack lengths are indicated at groups of data points); at 200 m/s impact velocities, fatigue cracking initiated at the base of indent site where the stress concentration is higher (compared to that at the crater rim) [5]. In contrast, at low applied stresses (and lower impact velocities), where FOD-induced microcracks were not formed or were much smaller, fatigue cracks were found to initiate away from the indent crater, in regions of relatively high tensile residual stresses. An example of fatigue crack initiation at low applied cyclic stresses ( σmax = 325 MPa, R = 0.1) in regions remote from the damage site in the interior of a KB specimen is shown in Figs. 5a and b for 300 m/s impacts of, respectively, 1 and 3.2 mm diameter shot. A further example (at σmax = 300 MPa, R = 0.1), where initiation occurred close to surface at the side face of the specimen is shown in Fig. 5c. In all cases, fatigue cracking initiated in regions of the high tensile residual stresses away from the indent; specifically, these stresses were ~300 MPa, based on synchrotron x-ray diffraction measurements [10] and numerical computations (Fig. 3). The tensile residual stresses also elevate the load ratio, when simply superimposed on the applied far-field fatigue loading. Based on FOD-related failures at applied cyclic stresses of 270 to 300 MPa (R = 0.1) and 250 MPa (R = 0.5), superimposing a measured tensile residual stress of ~300 MPa increases these R-ratios from 0.1 to ~0.5 and from 0.5 to ~0.7, assuming no relaxation. The corresponding increase in the maximum stress by this superposition is marked by arrows in the S-N diagram shown in Fig. 4. From these S-N results it is
RkJQdWJsaXNoZXIy MjM0NDE=