PSI - Issue 75

Florian Kalkowsky et al. / Procedia Structural Integrity 75 (2025) 581–592 Florian Kalkowsky et al. / Structural Integrity Procedia 00 (2019) 000 – 000

585

5

3. Experimental studies 3.1. Test program and test setup

The influencing parameters considered important by the authors on the fatigue behavior of bearing type connections should be investigated using the test program shown in Table 2. These include the material strength of the components on the one hand and the joint configuration via the geometric parameters ( 1 , 2 , 2 ) on the other. In addition, connections with blind rivets were also tested to investigate the influence of alternative fasteners. All holes in the components were manufactured by drilling, as alternative manufacturing processes such as punching, laser, oxy-fuel or plasma cutting can significantly influence the fatigue behavior. The description of their influence on fatigue strength as part of a design proposal is still the subject of research, see Holch et al. (2025); Kalkowsky et al. (2025). The hole clearance for the bolted connections were in line with the requirements from the EN 1090-2 (2024). In contrast, the holes for the blind rivets were drilled according to the manufacturer's recommendation. All specimens were double covered connections with a fatigue critical center plate. The fatigue tests on bearing type connections were carried out at CAL downwards the yield strength on the testing machines S INCOTEC P OWERSWING 100kN and W ALTER & B AI LFV 50 kN. In accordance with the test procedure specified in DIN 50100 (2022), the load level method resp. pearl string method for series with a small sample size was used. It should be noted that the DC’s acc. to EN 1993-1-9 (2010) can be applied independently from the mean stress level. This also assumes a high load ratio of = 0.5 when determining this resistance value. The authors are aware of this fact and it is considered in the evaluation of the test results using the synthetic S-N curves. Nevertheless, to cover the finite life region with fracture results, the fatigue tests were performed at = 0.1, since the static shear resistance of the fastener as well as the bearing capacity were the limiting values for the maximum cyclical load. Specimens that reached the number of cycles of 2 (Series I to IV) or 5 (Series V to X) million were evaluated as run outs. For a sufficient statistical validation of the results, at least ≥ 12 fracture results per series were enforced, which is in line with the requirements of FprEN 1993-1-9 (2024). For all other test series with a sample size of less than < 12, only the S-N curve was determined for a survival probability of 50 %. The test frequencies were between = 45 – 51 Hz for the S INCOTEC P OWERSWING 100kN and = 15 – 15 Hz for the W ALTER & B AI LFV 50 kN. Table 2. Test program and specimen geometry for fatigue tests on bearing type connections

t nom [mm]

d 0 [mm]

e 1 [mm] 3.0 d 0 3.0 d 0 3.0 d 0 1.5 d 0

e 2 [mm] 1.5 d 0 1.5 d 0 1.5 d 0 3.0 d 0 1.5 d 0 3.0 d 0 1.5 d 0

p 2 [mm]

X k (f u,obs ) [N/mm²]

m = var. [-]

K t,zd [-] 3.19 3.19 3.19 9.04 3.19 6.83 2.82 3.16 3.16 3.16

# data

Series

Steel grade

Centre plate geometry

I

5 5 3 5 8 8 8 2 2 2

13 13 13 13 18 18 18

S235JR S460MC S700MC S235JR

- - - - - - - - -

453 633 813 453 546 546 546 410 410 634

7.2 6.7 5.3 7.1 5.5 4.2 4.7

5 8

II

III IV

10

7

V

S335J2+N 3.0 d 0 S355J2+N 3.0 d 0 S355J2+N 3.0 d 0

14 14 13 16 15

VI

VII

2.5 d 0

VIII

6.7

S235JR S235JR S500MC

~3.0 d 0 ~1.5 d 0

10.8

IX

10 10

3.0 d 0 3.0 d 0

1.5 d 0 1.5 d 0

7.8 6.2

X 16 Series I - VII – M12 or M16 HV bolting assembly acc. to EN 14399 | Series VIII - X – Blind rivet M AGNA -L OK 6.4 mm and M ONOBOLT 10 mm

As the tensile strength of the component is an input parameter for the calculation method in accordance with the FKM-Guideline (2020), in addition to the fatigue tests the mechanical properties of the base materials were determined in tensile tests performed based on ISO 6892-1 (2020). A 5 % fractile value of the tensile strength k ( u,obs ) was determined by statistical evaluation of the = 3 results using the procedure from EN 1990 Annex D EN 1990 (2021).