The idea about what writing this month comes from a recent conference I attended, which focused on treatments and processes to improve the mechanical performance of components of cars. Shot peening was among the most widely discussed treatments, since it can be applied to most of the parts that needs an excellent fatigue strength. One of the speakers showed the results of a series of fatigue tests on samples made with a low alloy steel, which had the classical "hour-glass" shape, giving a unique minimum cross section where the fatigue crack is likely to appear and does not introduce any stress concentration, that is, it does not induce any notch effect. After having well explained the material and having said that the fatigue tests were reversed axial fatigue tests, the presenter concluded that after shot peening the fatigue limit of the samples changed (in better) for no more than 10%, which means that the shot peening application on that steel is not a good solution.

Indeed, I, to be sincere, have assisted to presentations like this more than once, and I am always surprised how a conclusion like this can be drawn.

It is well known that an hourglass sample under axial load has a uniform distribution of the stresses, and the stress value is the ratio of the applied force over the area of the cross section. Now, consider the compressive residual stresses introduced by shot peening in the surface layer of material: they will be beneficial and will improve the fatigue strength of the surface, but they cannot do anything at the core, where the residual stress field will be slightly in tension. Finally, the effect of shot peening will shift the crack initiation point from the surface to an inner part of the sample, probably in correspondence of an internal defect of the material, with a limited benefit (generally around 10%) for the fatigue limit. That's clear!

If we consider the same sample loaded in bending for a rotating bending fatigue test, the effect of shot peening will be tangible: the expected improvement of the fatigue limit with respect of the untreated parts should be about 20%, and the crack initiation point will remain on the free surface, apart particular cases.

Finally, if a notched part with a small fillet radius (or a hole or a groove), as usual in most of automotive machine elements, the result, after the application of shot peening, will be much more remarkable: the fatigue limit will be much higher than that of the unpeened parts, and it is not rare to improve it by more than 50%!

This is due to the stress gradient that characterizes smooth samples loaded in bending and notched samples, loaded in bending or also axially. In these cases, the effect of the residual stresses close to the surface remains valid while the core (where the residual stresses decrease and then become tensile) is much less stressed than the surface. And the more severe the stress gradient is, the greater will be the effect of shot peening.

Why testing the effectiveness of shot peening on smooth samples under axial fatigue when the component of interest has a different geometry? The result will be meaningless and cannot give any useful indication. Notched samples able to reproduce the stress profile and gradient from the surface to the core should be used.

This means that the optimization of shot peening of con-rods, crankshafts, springs, etc. should be done by designing samples able to reproduce the real stress field caused by the applied loads in the critical details.

...And this is what I would like to hear at the conferences.

Contributing Editor MFN and 

Full Professor of Technical University of Milan

20156 Milan, Italy

E-mail: mario@mfn.li