Keywords : shot peening, simulation, ultrasonic shot peening, nanocrystallisation, fatigue

1. Introduction

The research objectives of LASMIS (Laboratory of Mechanical Systems and Concurrent Engineering) are to propose new CAD/CAM tools for the designer in an industry with a concurrent engineering approach. Existing tools offer good solutions for shape design and strength evaluation via finite element computation; they offer only limited help for the complete optimisation of a mechanical part. Our concurrent engineering approach wishes to propose tools that will optimise the entire life of the mechanical component, from its shape design to its recycling possibilities via manufacturing processes and fatigue life evaluation. To reach this general objective we first concentrated our efforts to propose a computer tool for the optimisation of fatigue life using pre-stressing engineering [1]. Residual stresses whether coming from general manufacturing processes or pre stressing techniques greatly influence fatigue life. So we wish to propose a tool that will optimise the manufacturing parameters of a given pre-stressing technique as a function of the fatigue life specification. The research has been carried out on different techniques for residual stress evaluation: X-ray and neutron diffraction, incremental hole drilling, ultrasonic method, moiré interferometer method. Then, different models have been developed for the modelling of material processes including the simulation of the shot peening. Finally, the fatigue life prediction approach with the residual stress consideration has been developed. This paper summarizes the new progress carried out by LASMIS in the field of shot peening.

2. Development of the Shot Peening Process

2.1. Ultrasonic Shot Peening
The ultrasonic shot peening process (US) consists of impacting the surface of the part to be treated with spherical shot using high-power ultrasound. Like conventional shot peening (CS), it results in plastic deformation of the surface fibres thus inducing a compressive residual stress near the surface of the work piece, with the aim of improving the fatigue life of the component. The main differences between US and CS are as follows:
??The shot diameter is generally larger (between 0.4 and 3 mm) than that used in the conventional process,
??The shot velocity is random and lies within a range of 5 to 20 m.s-1,
??US shot is perfectly spherical, thus producing a smoother surface on the treated part than CS shot,
??The shot is harder (60 to 65 HRC) resulting in deeper shot-peened layers and less breakage of shot during treatment. This eliminates the need for new shot and shot screening.
??Only high strength shot such as 100C6 bearing shot is used in order to prevent breakage and to produce deep shot-peened layers.
To compare the influence of US and CS on fatigue life, fatigue tests were carried out on a steel steering ball joint used in the automotive industry [2]. The residual stress profiles without shot peening (WS), after conventional shot peening (CS) and after ultrasonic shot peening (US) are illustrated in figure 1. Figure 2 gives the results of the fatigue test. US improves the fatigue behaviour of the part to a more significant extent than CS, especially in the case of the lowest loading (520 daN). On the one hand, US produces a smoother surface than CS, which is beneficial for the fatigue life as a high level of roughness can increase crack initiation. On the other hand, the compressive stress level is higher in the case of ultrasonic shot peening (see figure 3). This study emphasises the possibility of improving the fatigue resistance of parts much more significantly by using US rather than CS, especially in the case of high-cycle fatigue loading.

2.2. Surface Mechanical Attrition Treatment (SMAT)
A new series of patented techniques has since been developed [3], namely surface mechanical attrition (SMA), to synthesize a nanostructured surface layer on bulk metallic materials. As a result of plastic deformation of the surface layer induced by mechanical attrition, the coarse-grained structure in the surface layer is refined to the nanometer scale without changing the chemical composition. The SMA technique has been successfully applied to various kinds of materials including pure metals, steels and Al, Ti, Ni based alloys, on which a nanostructured surface layer up to 50 µm thick has been obtained. A significant enhancement of the overall properties and performance of the materials has been observed after SMA treatment. Three categories of surface mechanical attrition (SMA) machines have been developed. The key issue with all these techniques is to severely and randomly plastify the surface of bulk materials. The first technique is based on the vibration of spherical shot using high power ultrasound. The second technique is based on the mechanical vibration of a reflecting chamber with shot ranging in diameter from several mm to 10 mm. The frequency is lower than that of an ultrasonic system, but the shot used can be much larger than those used in the previous system. It is also easier to introduce multidirectional movement of the reflecting chamber. Figure 3 gives a diagram of one possible configuration for this type of machine. Several new generations of machines are currently being developed.

3. Modelling of the Shot Peening Process
 
Finite element models can offer an in depth representation of the mechanical phenomena happening during the process modelled. On a fundamental scale, it enables one to better understand the process to be modelled, on a more practical scale, it can be used in an industrial context to validate treatment processes and understand the influence of process parameters.
In the model, shots of radius R are projected with a velocity V on the treated structure that generally has a complex geometry. Because the radius of curvature of the structure is always much larger than the radius of the shot, a semi-infinite volume can represent the shot-peened part. Mesh and elements have been optimised to guaranty computational efficiency. Figure 4 illustrates the mesh used for 3D computations and shows mesh displacement, plastic deformation and residual stress after nine impacts on an aluminium massif.

Although most existing models use a rigid shot for simplicity, it is a fact that during the industrial process, the spheres constituting the shot are elastically and plastically deformed. Then, the impact prints a weaker dent in the material and the residual stress profile is affected accordingly. It seems interesting to quantify these effects and we have proposed a model to study the influence of the parameters associated with the shot, that is to say their velocity, radius and yield stress [4]. The parameters used in the model have been chosen because they correspond to cases commonly found in industry for shot peening processes. AFNOR normalisation advises a shot hardness between 400 HV and 600 HV for standard applications. Five types of material’s parameters are considered for the shot including the perfectly rigid and perfectly elastic case. Several values of shot radius and velocity have been also considered (R = 0.15, 0.3, 0.45 mm and V = 20, 55, 80 m/s). The shot peened part is a steel massif.
Figure 5 presents a compilation of the residual stresses after one impact. The maximal value of the compressive stress obtained for each impact modelled is reported on the figure as a function of the shot’s velocity (the radius is also given) and for several shot’s materials. It is interesting to first look at the influence of the shot’s velocity and radius. Well known facts are confirmed here: increasing the velocity increases the maximal residual stress, whereas, the radius has very little influence on the maximal residual stress. There is an exception to this phenomenon: the shot with a yield stress of 1200 MPa impacting a steel solid generates a maximal residual stress that decreases with an increasing velocity. For this last case, the yield stresses of both materials are close (1200 MPa for the shot and 950 MPa for the steel) and it can be observed in the numerical results that an important part of the shot’s initial kinetic energy leads to the plastic deformation of the shot itself. The plastic deformation within the shot strongly increases with the velocity leading to a decreasing value of the maximal residual stress in the shot peened material.

4. Conclusion and Perspectives

Shot peening is now a popular process for increasing the material performance. We continue to develop the integrated tool of high cycle fatigue life simulation (Fatigue 3D). The 3D fatigue life can be calculated with residual stress and residual stress relaxation considerations. The experimental validation shows that our approach can predict with high precision the effect of shot peening on the fatigue life for a range between 100 thousands cycles to 10 millions cycles. For the process control, a new development is the use of a new ultrasonic technique for the through thickness residual stress determination using the multi frequency sensors system. This is a new NDE system for the residual stress analysis. In the field of process, we have shown some new extensions of the shot peening process for the generation of nanostructure, which produce some unique properties for the metallic and ceramic materials. This new kind of process based on the random ball motions will certainly extend the application fields of mechanical surface treatment.

5. References

1. Lu J. “Prestress Engineering of Structural Material: A Global Design Approach to the Residual Stress Problem”, Handbook of Residual Stress Formation and Distorsion of Steel, ASM, 2002, p. 11-26.
2. Retraint D., Garnier C., Guelorget B. and Lu J., “Improvement of the fatigue behaviour of an automotive part using a new mechanical treatment”, Materials Science Forum, Vol. 404-407, 2002, pp 463-468.
3. Lu J. and Lu K., Surface nanocrystallization (SNC) of materials, and its effect on mechanical behaviour, Encyclopedia Comprehensive Structural Integrity, Vol. 8 Chapter 14, pp495-528, Elsevier 2003
4. Rouhaud E., Deslaef D., “Influence of shots’ material on shot peening, a finite element model”, Materials Science Forum, Vol. 404-407, 2002, pp 153-158.

Authors: Delphine Retraint, Emmanuelle Rouhaud, Jian Lu
Laboratoire des Systèmes mécaniques et d’Ingénierie Simultanée (LASMIS)
Université de Technologie de Troyes
BP 2060-10010 Troyes Cedex France
Fax: 33.3.25 71 56 75
E-Mail: delphine.retraint@utt.fr