*3.4. Resonant Fatigue Resistance*

The presence of compressive RS itself does not insure an improvement in fatigue strength. The increase in surface roughness must be considered, because surface defects may promote initialization and propagation of cracks during dynamic loading. The fatigue test results are shown in Figure 16. It was found out that the maraging steel in a delivered state (maraging solution annealing, with hardness values of 32–37 hardness Rockwell scale C (HRC) and an *Rm* value of 1200 MPa) withstood the same number of fatigue cycles for crack initiation, which was precipitation-hardened, when the fatigue testing was performed at a bending moment of 60 N·<sup>m</sup> and a bending stress of 833 Mpa. When the bending stress was increased to 1082 Mpa, the precipitation-hardened maraging steel showed a 43% increase in crack initiation time (blue line in Figure 16). As can be observed, the LSP was proven to be successful in improving the fatigue resistance of precipitation-hardened maraging steel. The fatigue resistance was improved after every combination of laser SD and PD, which was tested within our research. Therefore, the negative e ffect of the increased surface roughness did not overcome the positive e ffect of the compressive RSs, present in the thin surface layer of the laser-peened material. According to our statistical analysis, we discovered that the selection of laser SD between 1.5 and 2.5 mm in combination with a PD of 900 or even 2500 cm<sup>−</sup><sup>2</sup> did not have a statistically significant effect on the fatigue resistance when comparing the LSP-treated specimens. The selection of laser processing parameters had a statistically significant e ffect on fatigue life only, when a laser SD of 2.0 mm was used ( *P* < 0.0001). The LSP with an SD of 2.0 mm and a PD of 1600 cm<sup>−</sup><sup>2</sup> increased the fatigue life of the precipitation-hardened maraging steel specimens from a range of 2–4 × 10<sup>4</sup> cycles to a range of 5–9 × 10<sup>5</sup> cycles. Therefore, the number of fatigue cycles, necessary for fatigue crack initiation, increased by 25 times. In some cases, we stopped the fatigue test after 10<sup>7</sup> cycles without crack initiation.

The implementation of the LSP technique can increase fatigue resistance, by generating compressive RSs and inducing strain hardening in most loaded surface areas, which consequently reduces the need for tool repair. Hence, when we succeed in increasing tool maintenance intervals (repair or change), the production cost is reduced.

**Figure 16.** Fatigue test results.

Further, the cyclic loading after the crack initiation phase causes crack propagation and leads to a continuous decrease of the resonant frequency until final fracture. Other researchers have also studied the e ffect of LSP on crack propagation behavior [24]. Within our research, we compared the behavior of the resonant frequency between the untreated and laser-peened specimens during the fatigue loading. *Metals* **2019**, *9*, 1271

As can be observed in Figure 17, the resonant frequency was decreasing more slowly, when the surface was treated with LSP. This finding indicated that laser peening not only extended the fatigue crack initiation time, but also reduced the crack propagation rate.

**Figure 17.** Resonant frequency behavior within the crack propagation phase during the fatigue bending: (**a**) MPH; (**b**) MPH-900-2.5; and (**c**) MPH-1600-2.0. The maximum bending moment was 78 N·m.

Figure 18a presents the failure locations on the specimens without LSP. The crack grew at the surface, because the LSP process slowed down thanks to the presence of the compressive RSs.

By laser peening with a 2.0 mm-diameter laser spot, the generated compressive RSs were in a range of 800–900 MPa, such that their positive effect could overcome the critical stress on the narrowed site in the middle of the fatigue specimen. It can be observed that the fatigue crack initiated at the edge of the LSP area (Figure 18c) and was not in the narrowest position in the middle of the specimen (Figure 18b).

When choosing the best laser peening parameters for increasing the fatigue life of a component, we must take into account a combination of the highest hardness and the highest compressive RS at a grea<sup>t</sup> depth (in our case, it is 1 mm). Laser spot size, in combination with laser power density and overlapping degree of laser pulses, affects the propagation of shock waves [26]. Small-diameter shock waves have different attenuation rates from those of large-diameter shock waves, when they go deep. In our case, we obtained the best combination of surface quality and material properties with the following LSP parameters: laser PPD = 8.9 GW·cm<sup>−</sup>2, laser PD = 1600 cm<sup>−</sup>2, laser SD = 2 mm, laser spot overlapping degree = 87%.

**Figure 18.** Fatigue crack and failure location on the specimen without LSP (**a**), the specimen with LSP for PD = 900 cm<sup>−</sup><sup>2</sup> and SD = 2.5 mm (**b**), and the specimen with LSP for PD = 1600 cm<sup>−</sup><sup>2</sup> and SD = 2.0 mm (**c**).
