*2.3. Resonant Fatigue Tests*

In order to obtain insight into the e ffects of LSP on the mechanical fatigue resistance of the chosen maraging steel, fatigue tests were carried out using a resonant testing machine, CRACKTRONIC (RUMUL, Neuhausen am Rheinfall, The Switzerland), which is a table model for dynamic bending load applications with testing frequencies between 40 and 300 Hz. The kinematic conditions allows for pure bending between the gripping heads. An electromagnetically driven resonator, built as a rotary oscillator, creates appropriate bending moments. The elastic modulus of a material and the specimen geometry have e ffects on the resonance frequency of the machine. The crack initiation and propagation in this unit reduce its cross-sectional area, which a ffects sti ffness reduction and consequently resonance frequency, as stated also in [23]. This machine measures resonance frequencies with a resolution of 0.01 Hz. In this research, the bending moment *M* was applied in a sinusoidal wave form at a stress ratio *R* of 0.1. We chose this stress ratio to provide and keep the upper surface of the specimen, where we expected crack initiation, in tensile conditions during fatigue tests. The bending frequency was around 114 Hz. We further reduced the middle of the specimen to ensure higher semicircular bending stresses in this area (Figure 2). In our resonant tests, we applied bending moments in a range between 60 and 78 N·m. The parameters of fatigue loading used in the experiment are presented in Table 4. According to our simulation, the maximum bending moment within a fatigue cycle generated a load of 1082 MPa in the most critical point of the specimen. The stress map during the bending is represented in Figure 3. Stress analysis was simulated with a software package SolidWorks (v23, 2015) according to the finite elements method (FEM) (3DEXPERIENCE, Vélizy-Villacoublay, France).


**Figure 2.** Principle of mechanical fatigue testing. The bending moment ( *M*) range: 60–78 N·m.

Figure 4 shows the change of the resonant frequency in dependence of the number of fatigue cycles, where *N*i represents an initiation period and *<sup>N</sup>*p indicates a propagation period. The resonant frequency, conditioned by the specimen's geometry, began to decrease, when the fatigue crack occurred. This event, which also separated the fatigue crack initiation phase and the fatigue crack propagation phase, was chosen as the moment of failure. We did not measure specimens' temperatures during the fatigue testing, and we did not notice any temperature increase in the specimens during and after the

test. The CRACKTRONIC is a compact testing device, so heat generated in the specimen during the fatigue loading could possibly be transferred to the resonant device.

**Figure 3.** Stress maps during the bending within the fatigue specimen: (**a**) side-view stress map; (**b**) top-view stress map. The maximum stress indicated in red was approximately 1000 MPa. The bending moment was 78 N·m.

**Figure 4.** Resonant frequency as a function of the number of fatigue cycles during the fatigue test.
