It is well known that residual stresses strongly affect the fatigue strength of machine elements; the effect is positive if they are compressive and negative if they are tensile. Since fatigue damage most times originates on the free surface of a structural element, different surface treatments were developed and adopted. Among these latter, shot peening is, probably, the luckiest and the most widely used, due to its applicability to general geometries, to the fact that it does not require high-tech devices and, most recently, for its low environmental impact.
We know also that automotive shot peening is very popular and ever more used to improve the fatigue strength of gears, crankshaft, welds, and other structural parts and engine components.
The way shot peening is performed is synthetically expressed by the Almen intensity, that is the residual arc height of a thin strip (the Almen strip), peened on one side only. This method, used all over the world, is really effective from an industrial point of view; it is fast, cheap and does not require a complex equipment. But it cannot give us direct and quantitative indication about the residual stresses induced by shot peening. This is a weak point.
Today we want to be sure about the quality of our treatment and about the entity of residual stresses induced by the treatment, to avoid unexpected failure and the related consequences.
But to know the residual stresses is not easy and their measurement complex.
Indeed, the measurement of residual stresses is a topic of technological interest from many years and many methods were proposed and validated, at least for some case of interest. The problem is that most of these methods are destructive, that is to say that they require to cut, partially or totally, the part of interest to relate the measured strain to the self-stresses acting before the cut is done. And this is not good if we wish to adopt residual stress measurement as a quality control method.
There are three methods that can be applied in the automotive industries for their reliability, accuracy and their generality: the hole drilling method, X-ray diffraction, and the Barkhausen noise technique.
In the hole drilling method, a strain gauge rosetta is applied on the surface of the part to be measured and a blind hole with a small diameter is practised in the centre of the strain gage rosetta: while the hole is done, the strain relaxation at location of the strain gages is measured and, by means of theoretical formulas, the result in terms residual stresses at the center of the rosetta is obtained. With this method, it is possible to appreciate in-depth stress gradients and the experimental equipment needed to perform the measurement is relatively cheap. And to obtain a complete in-depth profile is rather fast. Anyway, we can define it as a semi-destructive method, since we introduce a small hole that can prevent a subsequent utilization of the measured part.
The X-ray diffraction technique is a non- destructive method if applied to measure surface residual stresses. But it introduces surface damage if we need to perform in-depth measurements. In fact, due to the small penetration of X-ray in metals, it is necessary to remove small layers of material, possibly with an electro-polishing device, to prevent a strong alteration of the pre-existing residual stress state. X-ray diffraction angles are related to the residual stresses by a well-established theoretical basis, even if sometime the results are hardly interpretable due to preferential orientation of grains or other non-controllable factors. Beside, it requires an expensive instrument.
On the other hand, the Barkhausen noise technique is not based on any theoretical basis but on the experimental observation about the relation of this magnetic noise to the residual stress state. The problem is that the Barkhausen noise depends also on other factors and its application to measure the residual stress state requires a fine calibration on parts identical, from a micro-structural point of view, to the part of interest and with a well-known residual stress state (previously measured with another method). Nevertheless, it is fast and non-destructive for in-depth measurements and the equipment is cheaper than an X-ray diffractometer. Anyway, it requires dedicated probes for particular geometries, like the ones we are interested in for most cases. I know that this method is used for on-line quality control in luxury car companies.
To summarize, the measurement of residual stresses is still an important technological problem related to shot peening application in automotive companies. It is far from being completely solved and advancements are needed to increase its routine application.

Shot Peening in the Automotive Industry
by Mario Guagliano
Contributing Editor MFN and
Associate Professor of Technical University of Milan
20156 Milan, Italy
E-mail: mario@mfn.li

Author: Mario Guagliano