PSI - Issue 75

Jan Papuga et al. / Procedia Structural Integrity 75 (2025) 289–298 Author name / Structural Integrity Procedia (2025)

291

3

The issue of both concepts is that they describe the change in the stress across the evaluated space, but not the size of volume affected by high loads. It can be well explained on the example of plane bending and rotating bending load modes. Though the stress distribution is identical in one instant, the fatigue damage on the rotating bending specimen should be more significant and shorter fatigue life is expected simply because the volume of the part affected by high stresses is larger. This is the probable reason, why FKM-Guideline in v.6 [1] introduced a newer solution, in which the stress gradient analysis makes only a part of three concurrent effects. The other two are the size effect and the plasticity effect. A closer look at the analysis concept can be found in [1] i or in [2], where it is scrutinized in a smaller benchmark study. Within the master thesis of L. Serri [8], it was decided to check whether the integration of the concept of the critical volume only can provide results comparable to any of the methods presented on a small benchmark sets in [2], and whether the critical volume solution can represent both the stress gradient effect and the size effect.

Fig. 1. Three different ways to treat the notch effect via Theory of Critical Distance (TCD, green), Relative Stress Gradient (RSG, violet) or Nominal stress approach (nom, brown). K t designates stress concentration factor, K f notch factor and n fatigue factor. The meaning of the stress axis in the graph is related to the stress quantities depicted in the notch to the left from the graph. 3. Experiments Due to the lack of space in this conference paper, the basic set of experiments on unnotched full bar specimens from ČSN 411523 structural steel equivalent to S355J2 with active diameter of 8 mm, circumferential U- and V notches and circumferential fillet are referred to as documented in [2]. To make the comparison of various methods more demanding (and in agreement with rightfully addressed objections of reviewers of that paper), the original test set presented there was extended by more specimens as will be described below. Because all three original notched configurations led to specimens with similar K t factor of about 2.2 in push pull, it was decided to test one more notched configuration with more acute notch (a circumferential fillet with the notch root radius of 0.2 mm) and another one with a blunter notch (a circumferential U-notch with notch root radius of 5.0 mm). The differences in K t values for different variants of the tests can be observed in Table 1.

Table 1. Net stress concentration factors using von Mises stress for all load modes and all five types of notched specimens.

K t (von Mises)

V-notch R =1.3 mm

U-notch R =1.5 mm

Fillet R =0.7 mm

U-notch R =5 mm

Fillet R =0.2 mm

Load mode

Tension Torsion

1.99 1.43 1.64

1.89 1.39 1.56

2.10 1.48 1.81

1.30 1.14 1.18

3.34 2.06 2.76

Plane bending

In addition to those two new notched configurations, it was reasoned that the unnotched hollow specimen, which was used in [2] to describe the behaviour of the unnotched specimen, does not represent well enough the notched configurations due to its relatively thin wall, which can speed up crack growth phase compared to the notched