Section 6 Cold forming

6.1.1 The advantage of stroke peening or cold rolling of fillets is the compressive residual stresses introduced in the high-loaded area. Even though surface residual stresses can be determined by X-ray diffraction technique and subsurface residual stresses can be determined through neutron diffraction, the local fatigue strength is virtually non-assessable on that basis, since suitable and reliable correlation formulae are hardly known.

6.1.2 Therefore, the fatigue strength has to be determined by fatigue testing; see also Ch 2 Guidance for Evaluation of Fatigue Tests. Such testing is normally carried out as four-point bending, with a working stress ratio of R = - 1. From these results, the bending fatigue strength (surface- or subsurface-initiated depending on the manner of failure) can be determined and expressed as the representative fatigue strength for applied bending in the fillet.

6.1.3 In comparison to bending, the torsion fatigue strength in a fillet may differ considerably from the ratio √3 (utilised by the von Mises criterion). The forming-affected depth that is sufficient to prevent subsurface fatigue in bending, may still allow subsurface fatigue in torsion. Another possible reason for the difference in bending and torsion could be the extension of the highly stressed area.

6.1.4 The results obtained in a full size crank test can be applied to another crank size, provided that the base material (alloyed Q+T) is of the similar type and that the forming is done so as to obtain the similar level of compressive residual stresses at the surface as well as through the depth. This means that both the extension and the depth of the cold forming must be proportional to the fillet radius.

6.2.1 The fatigue strength obtained can be documented by means of full size crank tests or by empirical methods, if these are applied to be on the safe side. If both bending and torsion fatigue strengths have been investigated and differ from the ratio √3, the von Mises criterion should be excluded.

6.2.2 If only bending fatigue strength has been investigated, the torsional fatigue strength should be assessed conservatively. If the bending fatigue strength is concluded to be x per cent above the fatigue strength of the non-peened material, the torsional fatigue strength should not be assumed to be more than 2/3 of x per cent above that of the non-peened material.

6.2.3 As a result of the stroke peening process, the maximum compressive residual stress is found in the subsurface area. Therefore, depending on the fatigue testing load and the stress gradient, it is possible to have higher working stresses at the surface in comparison to the local fatigue strength of the surface. Because of this phenomenon, small cracks may appear during the fatigue testing, which will not be able to propagate in further load cycles and/or with further slight increases of the testing load because of the profile of the compressive residual stress. Put simply, the high compressive residual stresses below the surface ‘arrest’ small surface cracks.

This is illustrated in Figure 3.6.1 Working and residual stresses below the stroke-peened surface as gradient load 2. It should be noted that straight lines 1, 2 and 3 represent different possible load stress gradients.

Figure 3.6.1 Working and residual stresses below the stroke-peened surface

6.2.4 In fatigue testing with full-size crankshafts, these small “hairline cracks” should not be considered to be the failure crack. The crack that is technically the fatigue crack leading to failure, and that therefore shuts off the test-bench, should be considered for determination of the failure load level. This also applies if induction-hardened fillets are stroke-peened.

6.2.5 In order to improve the fatigue strength of induction-hardened fillets it is possible to apply the stroke peening process in the crankshafts’ fillets after they have been induction-hardened and tempered to the required surface hardness. If this is done, it might be necessary to adapt the stroke peening force to the hardness of the surface layer and not to the tensile strength of the base material. The effect on the fatigue strength of induction hardening and stroke peening the fillets shall be determined by a full size crankshaft test.

6.3.1 The fatigue strength can be obtained by means of full size crank tests or by empirical methods, if these are applied to be on the safe side. If both bending and torsion fatigue strengths have been investigated, and differ from the ratio √3, the von Mises criterion should be excluded.

6.3.2 If only bending fatigue strength has been investigated, the torsional fatigue strength should be assessed conservatively. If the bending fatigue strength is concluded to be x per cent above the fatigue strength of the non-rolled material, the torsional fatigue strength should not be assumed to be more than 2/3 of x per cent above that of the nonrolled material.