agreement with the value experimentally observed. According to these considerations, the following explanation for the failure can be given: i) fretting initiated cracks at the housing surface; ii) some cracks grew to a length which reduced the fatigue limit enough to make it equal to the local stress; iii) purely fatigue-driven crack growth was then possible and led to failure in a few cycles. An estimation of the number of cycles required for the initiation of fretting cracks was beyond the scope of this study, but given the limited number of parts involved in the failures and the very long service life (above 70000 running hours) compared to the short times required for a critical crack to propagate until rupture, it is reasonable to assume that it took many million cycles.

Two actions were undertaken to mitigate fretting damage: (a) increase of the bearing crush height, to allow for a higher radial pressure and therefore hinder sliding between the bearing shell and the housing, and (b) shot peening of the bearing housing, to improve the sliding resistance and reduce the sensitivity to defects.

The latter point deserves some more comments. The beneficial effect of shot peening on fatigue strength is well known, with many examples reported in the literature. This mechanical treatment has been employed with success also to enhance the fretting fatigue resistance, especially for aerospace alloys such as A7075-T6 and Ti6Al4V [5,6]. Experimental findings suggest that shot peening almost does not affect crack initiation, but effectively hinders fretting crack growth by its compressive residual stress layer [7]. With an Almen intensity of 12 A (0.3 mm) and the tensile strength in Table 2, the equation reported by Wang [8] gives a compressive depth of 0.296 mm, which is close to the depth of the longest fretting crack found on connecting rod caps. It is therefore expected that such compressive stresses do not allow fretting cracks to grow to a length which is critical to fatigue resistance.

As a parallel action, a non-destructive ultrasonic-based inspection protocol was set to monitor connecting rods which were thought to be susceptible to excessive fretting damage. A 45°-angled probe was employed for the measures and a calibration factor was introduced and further refined as the amount of available data grew. As ultrasounds allow collecting only two-dimensional information, a critical defect area was defined. There where the value was exceeded, immediate replacement of the part was carried out. This method was proven to be effective in detecting parts potentially prone to fatigue failure and was validated by destructively investigating two of them. Magnetic particles inspection confirmed the presence of a crack in the location predicted by ultrasonic testing. The cracks were then force opened to allow an accurate comparison between the expected and real fracture surface extension (Fig. 3). The protocol was thus validated as a way to predict the incipient failure with good enough accuracy. None of the connecting rods subject to the above mentioned changes were affected by further failures.

3. Cracks found on inspected connecting rod caps after forced opening.

References

[1] P.C. Paris, M.P. Gomez, W.E. Anderson, A rational analytic theory of fatigue, The Trend in Engineering 13 (1961) 9–14. [2] M.H. El Haddad, T.H. Topper, K.N. Smith, Prediction of non propagating cracks, Eng. Fract. Mech. 11 (1979) 573–584.

[3] B. Atzori, P. Lazzarin, G. Meneghetti, Fracture mechanics and notch sensitivity, Fatigue Fract. Eng. Mater. Struct. 26 (2003) 257–267.

[4] M. Beghini, L. Bertini, V. Fontanari, Stress intensity factors for an inclined edge crack in a semiplane, Eng. Fract. Mech. 62 (1999) 607–613. [5] G.H. Majzoobi, K. Azadikhan, J. Nemati, The effect of deep rolling and shot peening on fretting fatigue resistance of Aluminum-7075-T6,

Mater. Sci. Eng. A-Struct. 516 (2009) 235–247.

[6] S.A. Namjoshi, V.K. Jain, S. Mall, Effects of shot-peening on fretting-fatigue behaviour of Ti-6Al-4V, J. Eng. Mater.-T. ASME 124 (2002) 222–228.