PSI - Issue 79

Oleg Plekhov et al. / Procedia Structural Integrity 79 (2026) 168–175

173

in crystallographic lattice inter planar spacings while varying the angle ψ between the sample surface normal and the scattering vector of the X-ray beam. A key challenge when applying this method to curved surfaces lies in geometric distortions: tilting the sample to adjust ψ on convex su rfaces causes defocusing of the X-ray beam and systematic shifts in diffraction peak positions. These artifacts directly distort measured strains and substantially increase measurement errors compared to flat specimens. Residual stresses were quantified using an Empyrean PANalytical X-ray diffractometer. Measurements employed Cu- Kα radiation over a 2θ range of 136 - 143°, corresponding to diffraction from the (213) crystallographic plane. To minimize defocusing effects, a parallel beam mirror was used to collimate the primary beam. Figure 6 presents X-ray diffraction data showing the relationship between residual stress-induced strains and sin²ψ.

a

b

Fig. 6. (a) d- spacing evolution as a function of sin²ψ for: untreated specimen ; (b) LSP-treated specimen.

Data analysis revealed: weak compressive stresses (13 MPa) on the surface of the untreated cylindrical specimen, significantly higher compressive residual stresses (165 MPa) on the LSP-treated sample surface. This method qualitatively confirms the effectiveness of laser shock peening (LSP). As previously noted, measurement uncertainties associated with geometric aberrations and beam defocusing may be substantial for curved surfaces. 4. Fatigue tests Fatigue tests with flat specimens were carried out using a Bi-00-100 servohydraulic machine. Testing involved uniaxial cyclic deformation with a stress ratio of R = 0.1 and stress amplitude of 10 kN and 16.5 kN for LSP treated samples. Fatigue test results are presented in figure 7. LSP yielded a positive effect, significantly increasing the fatigue life of treated specimens identical loading conditions. An increase in the stress amplitude by 60% led to the specimens fatigue failure at the initial number of cycles.

Fig. 7. The result of fatigue tests on virgin specimens and LSP treated specimens