E-Archive


Vol. 4
September Issue
Year 2003
BACK

Science Update


in Vol. 4 - September Issue - Year 2003
Shot Peening, Fatigue and Aluminium Alloys – A difficult engineering problem



Author: Dr. Chris A. Rodopoulos has done research on a variety of fatigue related engineering problems, from offshore structures to aerospace components and for a variety of sponsors including NASA, Rolls-Royce, Airbus, Qinetiq, European Union, etc. In the last 7 years he has been involved with surface engineering, especially coating, shot peening, laser shock peening and ultrasonic impact treatment. He is the author or co-author of more than 60 scientific journal papers and the co-author of two books on fatigue.


Figure 1: Residual stress relaxation patterns for 2024-T351 at a maximum stress of 240 MPa, a stress ratio of R=0.1 for a typical Airbus dogbone push-pull specimen.


Figure 2: S-N response of treated and untreated 2024-T351 at a stress ratio of 0.1.


Figure 3: The effect of shot peening on crack initiation and crack growth rate for 2024-T351 subjected to a peak stress of 270 MPa at a stress ratio of 0.1.


Figure 4: The effect of surface roughness, as-treated Rtm=25


Figure 5: The effect of surface roughness (Rtm=38


Figure 6: Subsurface crack initiation. The position of the crack nucleus is closer to the micro-hardness saturation point-see figure 7.


Figure 7: Relaxation of work hardening in terms of fatigue life for 2024-T351 peened to an intensity of 0.14A subjected to a peak stress of 240 MPa with a stress ratio of 0.1.

Introduction

For many years, engineers considered the application of shot peening (SP) as the panacea of every fatigue and corrosion problem. As a result a multi-billion dollar industry emerged in, pretty much, every part of the world. SP has been used in a variety of applications starting from skin panels to landing gears to compression springs even to cutting blades for lawn mowers.

In the last ten years however, the history of SP started to divert. Several academic and industrial reports (the reader should consult the last 4 volumes of the International Conference on Shot Peening – www.shotpeening.org) have shown that the subject of SP and fatigue is characterised by controversy and in some cases unexpected failures.

SIRIUS has a long history on surface engineering and fatigue. Today our track record includes numerous research projects ranging from shot to laser shock peening and a variety of governmental and industrial sponsors, including the European Union, Airbus, Rolls-Royce, Hyundai Motor Company, BAE Systems, the Hellenic Aircraft Industry, etc.

In the article that follows, a number of problems, which emerged from the application of SP on aluminium panels on behalf of Airbus UK, are discussed.

Shot Peening and Fatigue – A marriage in desperate need for a counsellor

Shot peening is a simple technology, you take a ball, cut wire, oxide, etc accelerate it, drive it through some nozzle and hit the material. Or, to make things a bit more sophisticated, learn how to control the process by using digital sensors, controllers, different peening media, coverage, intensity, Almen strips, the whole lot. It is so simple that people do shot peening in their backyards. However, we must avoid generalisation. There are several well- established and responsible companies that really know what they are doing.
However, when it comes to shot peening and fatigue, things can turn out to be very complex. Fatigue, is by definition, a complex issue which most of the time goes beyond the simple Paul Paris representation, da/dN=CKm. Before we go any further, it is important to concentrate on three peening products that are most related to fatigue, i.e. compressive residual stresses; work hardening and surface roughness. 

Unfolding the problem

One of the most controversial problems of SP and fatigue is the modelling of the effect of the residual compressive stresses on fatigue life. Herein, there are four routes [1]. In the most simplistic way, an engineer can assume a mean stress effect or an effective stress ratio and modify the S-N or -N curve. The second route is to modify the stress intensity factor (K) due to the residual stresses, in terms of an effective Keff and integrate the Paul Paris equation. The third route is to consider the residual stress as an intrinsic stress. Here, there are two different approaches: a) we can directly model their action as a crack closure stress or b) we can model their effect on crack tip plasticity. In the latter this can be done by modifying the well know Newman’s model for plasticity induced crack closure. Despite the pros and cons of each approach, the engineer will face additional problems:

Figures 1,2 show the distribution of residual stresses as a function of depth for different life periods. From the results it is clear that in the case of aluminium alloys the compressive residual stresses do not saturate to a stable state but continue to relax with loading cycles. Further research has shown that the relaxation pattern depends on the applied stress level (faster at high stresses), the peening intensity and the material itself [2,3]. 

Thus, the modelling of the residual stresses is extremely difficult and should require an extensive statistical approach. However, the stability of the residual stresses is not the only problem when it comes to modelling. Figure 2 shows a typical S-N curve of peened (0.17A in mm) and un-treated 2024-T351. Close examination at 270 MPa reveals that there is a life improvement of approximately 230%. This is due to an increased period of crack initiation and slow crack growth. To provide an accurate life estimate, the engineer should be aware of the crack growth rates. Figure 3 shows a typical crack growth rate at 270 MPa. Comparison between Figure 2 and 3 reveals that approximately 68% of the improvement is due to crack initiation with the remaining 32% attributed to crack growth. From the above, it is easy to understand that modelling of fatigue life should be in two parts. With the first addressing crack initiation and the second crack propagation. The problem however amplifies if we take into account the effect of surface roughness. The definition surface roughness would introduce a stress concentration which is expected to reduce the fatigue life of engineering components due to premature crack initiation and accelerated near surface crack growth. However, in the case of shot peened surfaces, surface roughness could have two additional effects: a) it would neutralise a portion of the residual stresses. In other words, a portion of the residual stress would be consumed to counterbalance the increased stress concentration [4] and b) it will change the residual stress relaxation pattern as shown in Figure 4. Here it is important to understand that if the relaxation of the residual stress is severe and at the same time fast as in the case of high peening intensities, it is possible that the surface roughness effect would not be neutralised resulting in a crack growth rate which is even faster than that of the untreated material (see Figure 5). 

Another problem related to the fatigue behaviour of SP components is the effect of work hardening. Traditionally, work hardening was considered beneficial considering that changes in the dislocation density will amplify the resistance of the material to plastic deformation and thus would decelerate the speed of a growing crack. However, recent works indicated that work hardening could be responsible for a premature loss of ductility, strain softening and subsurface crack growth (see Figure 6) [5-7]. The latter could change the probability of damage detection (PDD) and hence cause expensive changes in the non-destructive inspections.  Unfortunately, our understanding of the effects of work hardening on fatigue is still in its infancy and it will be sometime before we are able to provide some concrete results. Here it is important to note that work hardening might also relax under the application of strain as shown in Figure 7.

Conclusions

From the previous text the following conclusion can be drawn:

a) The modelling of the fatigue life of shot peened components is very complex.

b) If shot peening is to be used  in design, the residual stress relaxation pattern should be known.

c) The phenomenology of the residual stress and work hardening relaxation is still to be addressed.

d) The removal of the surface roughness to improve the fatigue life of components could result in significant changes of which the designer should be fully aware. 

e) The relaxation of work hardening could be the reason behind subsurface crack initiation.

f) Subsurface crack initiation due to shot peening may require expensive and more complex non-destructive techniques.

References

1. Jose Solis Romero (2002) Optimisation of the shot peening process in terms of fatigue resistance, PhD Thesis, The University of Sheffield.
2. Jose Maria Ordieres (2003) Investigating the stability of the residual stresses induced by controlled shot peening on 7150-T651 aluminium alloy subjected to cyclic loading, MSc Thesis, The University of Sheffield.
3. John Werner Eichler (2003) Relaxation of residual stresses induced by shot peening in high strength aluminium alloys, MEng Thesis, The University of Sheffield.
4. S. Curtis, E. R. de los Rios, C. A. Rodopoulos, and A. Levers (2003) - Analysis of the effects of controlled shot peening on fatigue damage of high strength aluminium alloys. International Journal of fatigue, 25, P59-66, 2003. 




Author: Dr. Chris A. Rodopoulos
Dipl-Eng (Patras), MSc (Nottingham), PhD (Sheffield), Eur-Ing, MASME, MESIS, MSAE, MIAA, MIMechE.
Northern Aerospace Technology Exploitation Centre, Structural Integrity Research Institute of the University of Sheffield (SIRIUS)
Department of Mechanical Engineering
The University of Sheffield, Sheffield, S1 3JD.
Tel: +44.114-2227710
Fax:+44.114-2227890
E-mail: chris@Rodopoulos.freeserve.co.uk