Fatigue Performance of Thin Laser Butt Welds in HSLA Steel

metals Article Fatigue Performance of Thin Laser Butt Welds in HSLA Steel Patricio G. Riofrío1, Fernando Antunes 2,* , JoséFerreira 2, António Castanhola Batista 3 and Carlos Capela 2,4   Citation: Riofrío, P.G.; Antunes, F.; Ferreira, J.; Batista, A.C.; Capela, C. Fatigue Performance of Thin Laser Butt Welds in HSLA Steel. Metals 2021,11, 1499. https://doi.org/ 10.3390/met11101499 Academic Editor: Koh-ichi Sugimoto Received: 24 June 2021 Accepted: 31 July 2021 Published: 22 September 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1 DECEM, Departamento de Ciencias de la Energía y Mecánica, Universidad de las Fuerzas Armadas-ESPE, Av. General Rumiñahui S/N, Sangolquí171103, Ecuador; [email protected] 2CEMMPRE, Department of Mechanical Engineering, University of Coimbra, 3030-788 Coimbra, Portugal; [email protected] (J.F.); [email protected] (C.C.) 3CFisUC, Department of Physics, University of Coimbra, 3004-516 Coimbra, Portugal; [email protected] 4ESTG, Department of Mechanical Engineering, Instituto Politécnico de Leiria, Morro do Lena—Alto Vieiro, 2400-901 Leiria, Portugal *Correspondence: [email protected]; Tel.: +351-9347-98913 Abstract: This work is focused on understanding the significant factors affecting the fatigue strength of laser-welded butt joints in thin high-strength low-alloy (HSLA) steel. The effects of the weld profile, imperfections, hardness, and residual stresses were considered to explain the results found in the S-N curves of four welded series. The results showed acceptable fatigue strength although the welded series presented multiple-imperfections. The analysis of fatigue behavior at low stress levels through the stress-concentrating effect explained the influence of each factor on the S-N curves of the welded series. The fatigue limits of the welded series predicted through the stress-concentrating effect and by the relationship proposed by Murakami showed good agreement with the experimental results. Keywords: fatigue strength; laser butt weld; HSLA steel; fatigue limit prediction 1. Introduction High-strength steels have high fatigue resistance to crack initiation, as this damage mechanism decreases with the increase of the material’s tensile strength. The welding of these materials may be done using laser welding due to low heat inputs, small heat-affected zones, and less softening of the material. However, reductions of up to 40% in the fatigue limit, low ratios of the fatigue limit to the ultimate strength of the base metal (BM), and fatigue limits that do not exceed 175 MPa were reported in [ 1 – 3 ] when thin high strength steels were laser welded. The relevant facts that comprise the differences between welded joints and elements with notches or other defects, are the presence of crack-like imperfections and the changes introduced by the heat input (HI) into the BM. This has led to the fatigue initiation period being neglected [ 4 , 5 ] and to the application of the fracture mechanics approach, considering only the period of crack growth for the assessment of the fatigue life [ 6 ]. However, for small sized cracks and for low stress levels, the initiation stage is considered [7,8]. The conventional approach used to predict the fatigue limit in notched elements based on stress concentration and stress gradient models is also used in welded joints [ 9 ]. In [10,11], several traditional models of sensitivity to notches and new models were analyzed for the prediction of the fatigue limit of elements with small imperfections and for the application in welded joints. Based on fracture mechanics, Murakami [ 12 ] proposed an expression for fatigue limit predictions of materials containing small defects or cracks as a function of the hardness and the size of defects, and Åman et al. [ 13 ] modified the mentioned expression to consider the effect of the notch root on the small defects and suggested the application to the welded joints. Several factors affecting fatigue strength in weldments are highlighted in [ 4 , 6 , 14 ], however, they can be summarized in the following: weld quality, residual stresses, material, Metals 2021,11, 1499. https://doi.org/10.3390/met11101499 https://www.mdpi.com/journal/metals Metals 2021,11, 1499 2 of 22 and size effects. Although it has been shown that the weld quality control [ 15 ] or the use of post-weld treatments [ 16 – 19 ] in welded joints with high-strength steels increases the fatigue strength by reducing the severity of its inherent imperfections, this is recognized in design guides, allowing the increase of fatigue class curve (FAT) when the imperfections were verified with nondestructive testing (NDT) or when an improvement method was used [ 5 ]; there is no differentiation in the increase according to the mechanical strength of the materials. The foregoing has been explained due to the fact that the growth of cracks dominates the fatigue life of the welded joints and because the fatigue crack growth rate (FCGR) is insensitive to the mechanical strength of the material [ 20 ]. However, as counterpart, it was shown that when the weld quality is high, for high-strength steels, fatigue strength is positively correlated to mechanical strength [21]. Microstructural features such as phases and hardness are factors influencing the fatigue behavior of the welded joints. Due to the welding process, the microstructures of the heat-affect zone (HAZ), fusion zone (FZ), and BM are different and since, in most cases, the fatigue starts and paths are in the HAZ or FZ, the properties of these zones are important. Thus, the hardness is used instead of those of the BM, improving the results or fatigue predictions as shown by Kucharczyk et al. [ 22 ]. In [ 23 ], it is shown that the increase of the proportion of martensite in the microstructure decreases the FCGR and increases the crack growth threshold. In welded joints, the size effects refer to the decrease in fatigue strength when the thickness of the plates increases [ 4 ]. For this reason, fatigue strength reduction factors are applied when 25 mm of thickness is exceeded [ 5 ]. Although there is evidence that the opposite effect occurs for thin thicknesses [ 7 ], this is not considered in the design guides. The general effect of notches and residual stresses on fatigue strength of welded joints is well known. In various studies, their combined effects were studied and a relative effect was observed. Watanabe et al. [ 24 ] reported that the fatigue strength is dominated by stress concentration factor (SCF) when this is high and by the residual stresses when SCF is small. Harati et al. [ 25 ] concluded that the residual stresses have a larger influence than the weld toe geometry, meanwhile, Nguyen et al. [ 26 ] established that for tensile residual stresses of low magnitude or for compressive stresses, the effect of an undercut is dominant. Shiozaki et al. [ 27 ] reported that due to the equality of results in the S-N curves between the as-welded and heat-treated specimens, the effect of residual stress should not be considered. In almost all previous works, relatively thick plates and conventional welding processes were used. There are few studies on the main factors influencing fatigue strength and on the application of the fracture mechanics approach to the evaluation of fatigue limits of thin steel plates in laser welded joints. In the work [ 28 ], for butt joints of 3 mm thickness in Strenx ® 700MCE steel welded by laser, the weld bead geometry, the imperfections, and the quality level of the welds according to the ISO 13919-1 [ 29 ] standard were determined. The fatigue strength of the mentioned welded joints was evaluated in this study. For this, local properties such as hardness and residual stresses in the HAZ and FZ and a more realistic weld profile will be considered. The actual profile of the weld beads, including imperfections, is completely modeled to determine the SCFs by FEM. The residual stresses are measured by X-ray technique. Through the failure analysis of fractured surfaces and using the stress-concentrating effect, the influence of the main factors on fatigue behavior at low stress levels is explained. Predictions of the fatigue limits of the welded series are also rendered both by the stress-concentrating effect and by the relationship proposed by Murakami. 2. Material and Procedures 2.1. Material and Laser Welding Butt weld joints were created with 3 mm thick high-strength low-alloy (HSLA) steel plate Strenx ® 700MCE [ 30 ]. This steel has the chemical composition and tensile mechanical properties shown in Tables 1and 2[ 31 ], respectively. It presents a fine-grain ferritic-bainitic Metals 2021,11, 1499 3 of 22 microstructure. The yield stress is close to the tensile strength, which indicates that this material keeps the elastic behavior almost until final fracture. Table 1. Base metal chemical composition (wt %). C Mn Si P S Cr V Nb Ni Cu Al Mo Ti Co Fe 0.07 1.69 0.01 0.012 0.006 0.03 0.02 0.046 0.04 0.011 0.044 0.016 0.117 0.016 balance Table 2. Mechanical properties of the base metal. Yield Strength (MPa) Tensile Strength (MPa) Elongation (%) 808 838 15.0 Five series were welded considering the welding parameters shown in Table 3. The S1, S2, S3, and S4 series are single-welded joints while the S5 series is double-welded joint. For this last series, the welding speed of the weld pass on the top side was different for each sample (1.75, 1.80, 1.90, 1.95, and 2.00 m/min), while the weld pass on the bottom side was constant for all samples (2.50 m/min). These parameters were selected based on a previous study of the effect of parameters on defects and tensile properties [ 31 ]. The heat inputs used in welding were defined in order to achieve different values of hardness in the welded series, but simultaneously maintaining small HAZs, low softening, and tensile mechanical properties similar to the BM, as shown in [ 31 ]. A disk laser equipment Trumpf TruDisk 2000 with: laser maximum output, 2000 W; beam wavelength, 1020 nm; beam parameter product, 2 mm-mrad; and fiber diameter, 200 µ m was used in continuous mode. Argon (99.996%) shielding gas was supplied only on the top side of the weld at a flow rate of 20 L/min. Figure 1a shows the size of plates used in the welding process. Table 3. Welding parameters used in the experimental work. Series Laser Power (kW) Welding Speed (m/min) Heat Input (J/mm) S1 2.00 1.60 75.0 S2 1.75 1.60 65.6 S3 2.00 2.00 60.0 S4 1.75 2.00 52.5 top side weld pass 1.75 1.75–2.00 52.5–60.0 S5 bottom side weld pass 1.25 2.50 30.0 In work [ 28 ], according to the welding quality standard ISO 13919-1, a B quality level for the S1, S2, and S3 welded series and a D quality level for the S5 welded series, were reported and according to the imperfection sizes reported, these can be considered as cracklike imperfections (see the imperfections sizes in Table 6). In the aforementioned work, the imperfections were measured using a profilometer with a pitch of 1 µ m and a microscope equipped with digital micro-meters. Since the S4 series did not reach full penetration, it will be omitted in the next sections. Therefore, this series defines approximately the lower boundary of the heat inputs in order to obtain full penetration of the welding in the 3 mm thick specimens. Metals 2021,11, 1499 4 of 22 Metals 2021, 11, x FOR PEER REVIEW 3 of 22 2. Material and Procedures 2.1. Material and Laser Welding Butt weld joints were created with 3 mm thick high-strength low-alloy (HSLA) steel plate Strenx® 700MCE [30]. This steel has the chemical composition and tensile mechanical properties shown in Tables 1 and 2 [31], respectively. It presents a fine-grain ferritic-bainitic microstructure. The yield stress is close to the tensile strength, which indicates that this material keeps the elastic behavior almost until final fracture. Table 1. Base metal chemical composition (wt %). C Mn Si P S Cr V Nb Ni Cu Al Mo Ti Co Fe 0.07 1.69 0.01 0.012 0.006 0.03 0.02 0.046 0.04 0.011 0.044 0.016 0.117 0.016 balance Table 2. Mechanical properties of the base metal. Yield Strength (MPa) Tensile Strength (MPa) Elongation (%) 808 838 15.0 Five series were welded considering the welding parameters shown in Table 3. The S1, S2, S3, and S4 series are single-welded joints while the S5 series is double-welded joint. For this last series, the welding speed of the weld pass on the top side was different for each sample (1.75, 1.80, 1.90, 1.95, and 2.00 m/min), while the weld pass on the bottom side was constant for all samples (2.50 m/min). These parameters were selected based on a previous study of the effect of parameters on defects and tensile properties [31]. The heat inputs used in welding were defined in order to achieve different values of hardness in the welded series, but simultaneously maintaining small HAZs, low softening, and tensile mechanical properties similar to the BM, as shown in [31]. A disk laser equipment Trumpf TruDisk 2000 with: laser maximum output, 2000 W; beam wavelength, 1020 nm; beam parameter product, 2 mm-mrad; and fiber diameter, 200 µm was used in continuous mode. Argon (99.996%) shielding gas was supplied only on the top side of the weld at a flow rate of 20 L/min. Figure 1a shows the size of plates used in the welding process. Figure 1. (a) Sketch of plates (dimensions in mm) for the welding and location of specimens, (b) dimensions (in mm) of specimens used in fatigue tests. Figure 1. ( a ) Sketch of plates (dimensions in mm) for the welding and location of specimens, ( b ) dimensions (in mm) of specimens used in fatigue tests. 2.2. Fatigue Testing and Analysis The fatigue specimens were prepared and tested according to the guidelines of the ASTM E466-15 standard [ 32 ]. Figure 1a illustrates the location of the specimens in the welded plates, which were obtained by laser cut, meanwhile, Figure 1b shows specimens’ dimensions. The fatigue specimens were tested in three conditions: not-welded (BM), as-welded, and bottom side-removed. For this last condition, the excess weld metal and undercuts were progressively eliminated in order to obtain a smooth transition between the weld and the plate, thus minimizing the stress concentration. The specimens were carefully ground to a final polish with sandpaper # 220, particularly the edges and the welding surfaces. The total elimination of the undercuts was verified using the optical microscope with 20×magnification power. The testing was performed using DARTEC INSTRON servo-hydraulic machine. The fatigue specimens were axially loaded with constant amplitude at a stress ratio R = 0 and at a frequency of 15 Hz. The specimens were tested up to the final fracture at various stress levels; however, when at a certain stress level in which the life of the specimens exceeded 2 × 10 6 cycles, the tests were stopped (run-outs). After the fracture of the specimens, their surfaces were analyzed with a stereo-microscope in order to determine the site, start, imperfection, path, and shape of the fatigue surface. A power law relationship was assumed for the S-N curves: log (N) = log (C) −m log (∆σ), (1) where the nominal stress range ∆σ is the independent variable, the number of cycles N at which the failure occurs is the dependent variable and C, m are parameters that characterize the curves. Through linear regression, the S-N data were analyzed according to the ASTM E739-10 standard [ 33 ] to determine the parameters (C, m) of the S-N relationships, the median S-N curve, and its 95% confidence band. The testing of adequacy of the linear model was also verified according to the mentioned standard. For comparison, the FAT100 curve (slope m = 3, ∆σ = 100 MPa at two million cycles) was also plotted in the S-N graphs. This FAT100 curve, suggested in the fatigue design recommendations of the International Institute of Welding (IIW) for butt joints when imperfections (undercuts and porosity) are evaluated [ 5 ], was chosen because its requirements closely match (differs in the thickness Metals 2021,11, 1499 5 of 22 and the porosity of one of the series) the imperfection sizes and quality level found in welded series, as reported in [28]. 2.3. Measurement of Residual Stresses The residual stress analysis was performed by X-ray diffraction using Proto iXRD equipment in the longitudinal and transverse directions of the laser welds and on both sides (top and bottom surfaces) of the specimens. Lattice deformations of the {211} diffraction planes (2 θ≈ 156 ◦ ) were measured using Cr-K α X-ray radiation, with 11 β angles in the range ± 30 ◦ (22 ψ angles in the range ± 42 ◦ ), an acquisition time of 30 s by peak, and ± 2 ◦ oscillations in ψ . An aperture length of 0.5 mm × 3 mm in the transverse and longitudinal directions, respectively, was used. The residual stresses were evaluated with an elliptical regression of sin 2ψ using the X-ray elastic constants of 5.83 × 10 −6 (MPa) −1 for 1 2S{211} 2 and − 1.28 × 10 −6 (MPa) −1 for S{211} 1 . For the analyzed material and considering the radiation used, the average penetration depth of the X-rays was about 5 µm. The measurement of the residual stresses was performed in the middle of the width of the fatigue specimen at points located along the horizontal line (hl), see Figure 1b. A 32 mm extension was covered with a separation between points that was progressively reduced when approaching the welded zone, these separations were: 4 mm, 2 mm, 1 mm, and 0.5 mm. Figure 1b shows the directions of the longitudinal (L) and transverse (T) residual stresses. 3. Results and Discussion 3.1. S-N Curves Figure 2shows the S-N experimental results of both the welded series and the BM. The mean and the 95% confidence band curves of the BM are also presented along with FAT100 curve, which is a reference for welded joints. The BM shows a linear behavior in log-log scale with small scatter. The nominal stress range varied between 600 and 800 MPa, which corresponds to 74% and 99% of the material’s yield stress, respectively. The modifications introduced by laser welding on microstructure significantly changed the position of the S-N curves. A significant decrease of stress range for the same fatigue life is evident, namely for lower load ranges, i.e., for longer fatigue lives. The scatter increased significantly and comparatively to the BM. Both the S-N curves of the BM and of the welded joints presented the characteristic “knee point” that identifies the stress level of the fatigue limit, which, in this work, will be assumed as the fatigue strength of run-outs at two million cycles. Eventually, the series S2 may be considered to have a higher limit. The FAT100 curve is clearly below the experimental results obtained for the welded series, which is a good indication for the quality of laser welding. Figure 3presents the results for each series and a photo-macrograph showing the typical cross-section of the weld bead. The stress levels corresponding to the fatigue limit are also indicated. The dispersion of the entire data set is reduced when each series is considered separately, and the dispersion of each series is similar to that of the BM. In the case of the S5 series, because two specimens (named as S5-F9) showed considerably higher fatigue strength than the rest of specimens, they were not considered within the S5 series. The FAT100 is clearly below the experimental results of all the series. Table 4presents log C and m parameters of S-N curves. There is a limited influence of welded series on these values, with the exception of S2 series. The m values are close to FAT100 curve slope. The relatively high m value obtained for the BM indicates that the initiation is dominant over the propagation. Conversely, the relatively low value of m for the welded joints indicates the predominance of propagation over initiation. The fatigue limits shown in Table 4, which were assumed to be the maximum stress range values reached by the specimens that exceeded two million cycles, are between 30% and 57% of the fatigue limit of the BM. As can be seen in Figure 3, all series had a fatigue life over the FAT100 curve. From the results obtained, the confidence F-test for all welded series and BM did not reject the hypothesis of linear model. Metals 2021,11, 1499 6 of 22 Metals 2021, 11, x FOR PEER REVIEW 6 of 22 Figure 2. S-N curves for all welded series and BM. Figure 3. S-N curves of each welded series. Figure 2. S-N curves for all welded series and BM. Metals 2021, 11, x FOR PEER REVIEW 6 of 22 Figure 2. S-N curves for all welded series and BM. Figure 3. S-N curves of each welded series. Figure 3. S-N curves of each welded series. Metals 2021,11, 1499 7 of 22 Table 4. S-N curves parameters and fatigue limits. Series Linear Model F Test F0.95 Fcalculated log C m Fatigue Limits (MPa) S1 6.59 0.27 (not rejected) 13.46 3.23 180 S2 9.28 7.86 (not rejected) 16.21 4.25 270 S3 3.97 1.62 (not rejected) 13.25 3.12 210 S5 3.69 1.57 (not rejected) 13.47 3.21 215 (340) 1 BM 3.70 0.44 (not rejected) 30.64 9.00 600 1The value 340 MPa corresponds to S5-F9 specimens. 3.2. Fatigue Failure Modes Based on the analysis of the fractured surfaces of all the specimens, mainly four failure modes were found in the welded series. These are designated as FM1, FM2, FM3, and FM4. Table 5presents the principal features of each failure mode: the series in which the failure mode was present; the site, imperfection, and side of the weld bead where fatigue started; details of the imperfections; the shape of the fractured surface; and remarks. Table 5includes representative photographs showing the fracture surfaces, the fatigue initiation sites, and details of the imperfections that caused the fatigue. For each mode, the photographs of two fractured specimens are presented with the exception of the FM4 mode, for which only the photographs are presented for one specimen. Table 5. Failure modes features. Failure Mode (Series) Fatigue Start Site-Imperfection (Side) Details Approximate Shape of Fractured Surface Remarks FM1 (S1, S3) root-undercuts (bottom side) (a) long and deep undercuts, one or two very close (b) several undercuts across the width (a) big semi-elliptical (b) tending to rectangular by coalescence of small semi-elliptical in some cases, microdefects close to undercuts Metals 2021, 11, x FOR PEER REVIEW 7 of 22 Table 4 presents log C and m parameters of S-N curves. There is a limited influence of welded series on these values, with the exception of S2 series. The m values are close to FAT100 curve slope. The relatively high m value obtained for the BM indicates that the initiation is dominant over the propagation. Conversely, the relatively low value of m for the welded joints indicates the predominance of propagation over initiation. The fatigue limits shown in Table 4, which were assumed to be the maximum stress range values reached by the specimens that exceeded two million cycles, are between 30% and 57% of the fatigue limit of the BM. As can be seen in Figure 3, all series had a fatigue life over the FAT100 curve. From the results obtained, the confidence F-test for all welded series and BM did not reject the hypothesis of linear model. Table 4. S-N curves parameters and fatigue limits. Series Linear Model F Test F 0.95 F calculated log C m Fatigue Limits (MPa) S1 6.59 0.27 (not rejected) 13.46 3.23 180 S2 9.28 7.86 (not rejected) 16.21 4.25 270 S3 3.97 1.62 (not rejected) 13.25 3.12 210 S5 3.69 1.57 (not rejected) 13.47 3.21 215 (340) 1 BM 3.70 0.44 (not rejected) 30.64 9.00 600 1 The value 340 MPa corresponds to S5-F9 specimens. 3.2. Fatigue Failure Modes Based on the analysis of the fractured surfaces of all the specimens, mainly four failure modes were found in the welded series. These are designated as FM1, FM2, FM3, and FM4. Table 5 presents the principal features of each failure mode: the series in which the failure mode was present; the site, imperfection, and side of the weld bead where fatigue started; details of the imperfections; the shape of the fractured surface; and remarks. Table 5 includes representative photographs showing the fracture surfaces, the fatigue initiation sites, and details of the imperfections that caused the fatigue. For each mode, the photographs of two fractured specimens are presented with the exception of the FM4 mode, for which only the photographs are presented for one specimen. Table 5. Failure modes features. Failure Mode (Series) Fatigue Start Site-Imperfection (Side) Details Approximate Shape of Fractured Surface Remarks FM1 (S1, S3) root-undercuts (bottom side) (a) long and deep undercuts, one or two very close (b) several undercuts across the width (a) big semi-elliptical (b) tending to rectangular by coalescence of small semi-elliptical in some cases, microdefects close to undercuts Metals 2021, 11, x FOR PEER REVIEW 8 of 22 FM2 (S2, S1, S3) root-excess weld (bottom side) multiple starts in micro-imperfections (pores, cavities, shallow crack-like) across the width tending to a rectangular stripe across the width semi-elliptical forms were not observed in weld root FM3 (S5) (S1,S2,S3) face-underfill (bottom side) (top side) multiple starts in waves and sharp ripples with microdefects small semi-ellipses or tending to a big semi-ellipse by coalescence in many cases, underfill at border Metals 2021,11, 1499 8 of 22 Table 5. Cont. Failure Mode (Series) Fatigue Start Site-Imperfection (Side) Details Approximate Shape of Fractured Surface Remarks FM2 (S2, S1, S3) root-excess weld (bottom side) multiple starts in micro-imperfections (pores, cavities, shallow crack-like) across the width tending to a rectangular stripe across the width semi-elliptical forms were not observed in weld root Metals 2021, 11, x FOR PEER REVIEW 8 of 22 FM2 (S2, S1, S3) root-excess weld (bottom side) multiple starts in micro-imperfections (pores, cavities, shallow crack-like) across the width tending to a rectangular stripe across the width semi-elliptical forms were not observed in weld root FM3 (S5) (S1,S2,S3) face-underfill (bottom side) (top side) multiple starts in waves and sharp ripples with microdefects small semi-ellipses or tending to a big semi-ellipse by coalescence in many cases, underfill at border Metals 2021, 11, x FOR PEER REVIEW 8 of 22 FM2 (S2, S1, S3) root-excess weld (bottom side) multiple starts in micro-imperfections (pores, cavities, shallow crack-like) across the width tending to a rectangular stripe across the width semi-elliptical forms were not observed in weld root FM3 (S5) (S1,S2,S3) face-underfill (bottom side) (top side) multiple starts in waves and sharp ripples with microdefects small semi-ellipses or tending to a big semi-ellipse by coalescence in many cases, underfill at border FM3 (S5) (S1, S2, S3) face-underfill (bottom side) (top side) multiple starts in waves and sharp ripples with microdefects small semi-ellipses or tending to a big semi-ellipse by coalescence in many cases, underfill at border Metals 2021, 11, x FOR PEER REVIEW 8 of 22 FM2 (S2, S1, S3) root-excess weld (bottom side) multiple starts in micro-imperfections (pores, cavities, shallow crack-like) across the width tending to a rectangular stripe across the width semi-elliptical forms were not observed in weld root FM3 (S5) (S1,S2,S3) face-underfill (bottom side) (top side) multiple starts in waves and sharp ripples with microdefects small semi-ellipses or tending to a big semi-ellipse by coalescence in many cases, underfill at border Metals 2021, 11, x FOR PEER REVIEW 9 of 22 FM4 (S5, S1, S2, S3) center-porosity (lateral side) single pore or several on or near the surface approaching the middle of a semi-elliptical in some cases influenced by underfill close to corner The FM1 failure mode was due to the undercuts at weld root of the S1 and S3 series; there was generally more than one undercut that generated semi-elliptical cracks. In the FM2 failure mode, multiple imperfections along the weld root caused failures, especially in the S2 series. In these cases, the presence of semi-elliptical cracks was not evident. The FM3 failure mode was present in practically all the welded series and the fatigue starts were located in the underfill and the ripples. The ripples corresponding to the S5 series were smaller; see Table 5. The FM4 failure mode refers to failure due to the presence of pores that were present in all series but in greater numbers in the S5 series. The pores caused fatigue failures at the lateral side of the specimens generally when they were located near or on the surface. This failure can also be influenced by the underfill at the border of the element, as illustrated in Table 5. It should be noted that in approximately 44% of the fractured specimens, at least two failure modes were observed (as illustrated in FM2 mode, S3 series in Table 5), however, in most cases, only one mode was decisive to fatigue failure. The fatigue path was generally straight from the weld root through thickness to the underfill or vice versa. However, in certain cases, the path had inclined parts due to the starts of fatigue from different planes by the presence of porosity, or in the final stages of the fracture, due to deviations to the soft material where greater deformations were presented. Figure 4 presents the failure modes observed in each specimen of the different series. Each series had a main failure mode that prevailed over the others, and all series presented at least two main and two additional failure modes. As can be seen, in S1 and S3 series the FM1 mode prevailed, in S2 series the FM2 mode is more relevant, while in S5 series the FM3 mode is dominant. The above is strongly related to the corresponding presence of imperfections in each weld bead; see Table 5. Figure 4 also illustrates, in the sketches on the cross-sections of the weld beads, the starts and paths followed by the four failure modes and that all the fractures in the specimens occurred within the FZ and HAZ. Considering the starts on underfill and the starts on excess weld and undercuts, it was observed that the last dominated in S1, S2, and S3 series while the first dominated in S5 series. The FM4 mode, which corresponds to porosity, caused few starts in all series, Metals 2021,11, 1499 9 of 22 Table 5. Cont. Failure Mode (Series) Fatigue Start Site-Imperfection (Side) Details Approximate Shape of Fractured Surface Remarks FM4 (S5, S1, S2, S3) center-porosity (lateral side) single pore or several on or near the surface approaching the middle of a semi-elliptical in some cases influenced by underfill close to corner Metals 2021, 11, x FOR PEER REVIEW 9 of 22 FM4 (S5, S1, S2, S3) center-porosity (lateral side) single pore or several on or near the surface approaching the middle of a semi-elliptical in some cases influenced by underfill close to corner The FM1 failure mode was due to the undercuts at weld root of the S1 and S3 series; there was generally more than one undercut that generated semi-elliptical cracks. In the FM2 failure mode, multiple imperfections along the weld root caused failures, especially in the S2 series. In these cases, the presence of semi-elliptical cracks was not evident. The FM3 failure mode was present in practically all the welded series and the fatigue starts were located in the underfill and the ripples. The ripples corresponding to the S5 series were smaller; see Table 5. The FM4 failure mode refers to failure due to the presence of pores that were present in all series but in greater numbers in the S5 series. The pores caused fatigue failures at the lateral side of the specimens generally when they were located near or on the surface. This failure can also be influenced by the underfill at the border of the element, as illustrated in Table 5. It should be noted that in approximately 44% of the fractured specimens, at least two failure modes were observed (as illustrated in FM2 mode, S3 series in Table 5), however, in most cases, only one mode was decisive to fatigue failure. The fatigue path was generally straight from the weld root through thickness to the underfill or vice versa. However, in certain cases, the path had inclined parts due to the starts of fatigue from different planes by the presence of porosity, or in the final stages of the fracture, due to deviations to the soft material where greater deformations were presented. Figure 4 presents the failure modes observed in each specimen of the different series. Each series had a main failure mode that prevailed over the others, and all series presented at least two main and two additional failure modes. As can be seen, in S1 and S3 series the FM1 mode prevailed, in S2 series the FM2 mode is more relevant, while in S5 series the FM3 mode is dominant. The above is strongly related to the corresponding presence of imperfections in each weld bead; see Table 5. Figure 4 also illustrates, in the sketches on the cross-sections of the weld beads, the starts and paths followed by the four failure modes and that all the fractures in the specimens occurred within the FZ and HAZ. Considering the starts on underfill and the starts on excess weld and undercuts, it was observed that the last dominated in S1, S2, and S3 series while the first dominated in S5 series. The FM4 mode, which corresponds to porosity, caused few starts in all series, The FM1 failure mode was due to the undercuts at weld root of the S1 and S3 series; there was generally more than one undercut that generated semi-elliptical cracks. In the FM2 failure mode, multiple imperfections along the weld root caused failures, especially in the S2 series. In these cases, the presence of semi-elliptical cracks was not evident. The FM3 failure mode was present in practically all the welded series and the fatigue starts were located in the underfill and the ripples. The ripples corresponding to the S5 series were smaller; see Table 5. The FM4 failure mode refers to failure due to the presence of pores that were present in all series but in greater numbers in the S5 series. The pores caused fatigue failures at the lateral side of the specimens generally when they were located near or on the surface. This failure can also be influenced by the underfill at the border of the element, as illustrated in Table 5. It should be noted that in approximately 44% of the fractured specimens, at least two failure modes were observed (as illustrated in FM2 mode, S3 series in Table 5), however, in most cases, only one mode was decisive to fatigue failure. The fatigue path was generally straight from the weld root through thickness to the underfill or vice versa. However, in certain cases, the path had inclined parts due to the starts of fatigue from different planes by the presence of porosity, or in the final stages of the fracture, due to deviations to the soft material where greater deformations were presented. Figure 4presents the failure modes observed in each specimen of the different series. Each series had a main failure mode that prevailed over the others, and all series presented at least two main and two additional failure modes. As can be seen, in S1 and S3 series the FM1 mode prevailed, in S2 series the FM2 mode is more relevant, while in S5 series the FM3 mode is dominant. The above is strongly related to the corresponding presence of imperfections in each weld bead; see Table 5. Figure 4also illustrates, in the sketches on the cross-sections of the weld beads, the starts and paths followed by the four failure modes and that all the fractures in the specimens occurred within the FZ and HAZ. Considering the starts on underfill and the starts on excess weld and undercuts, it was observed that the last dominated in S1, S2, and S3 series while the first dominated in S5 series. The FM4 mode, which corresponds to porosity, caused few starts in all series, however, in S5 series it increased as a consequence of its highest percentage of porosity (3.5%) [28]. 3.3. Effect of Weld Bead Geometry and Imperfections In order to evaluate the effect exerted by the weld profile and geometrical imperfections on the fatigue behavior, the stress concentration factor (SCF), K t , corresponding to the top and bottom sides of each welded series were determined by the finite element method Metals 2021,11, 1499 16 of 22 Metals 2021, 11, x FOR PEER REVIEW 16 of 22 Figure 11. Residual stresses profiles in fatigue specimens. 3.5.3. Fatigue Limit Evaluation In previous sections, it was shown that for low stress levels, the approach of reducing fatigue strength through stress-concentrating effect can be used. To consider the effect of the mean stress in the analysis of the fatigue strength at two million cycles of the welded series, due to its simplicity, the modified Goodman line will be used according to the expression: 𝐾 𝜎 𝑆+𝐾 𝜎 𝑆 =1 𝑛 (2) where 𝜎 and 𝜎 are the nominal alternating and mean stresses applied and both are equal to ∆σ/2 (according to stress ratio R = 0); 𝐾 , is the effective stress concentration factor calculated using the relationship of sensitivity to notch proposed by Peterson [37]; 𝑆 and 𝑆 are the ultimate strength and the fatigue limit of the materials that will be estimated for each welded series and BM through the hardness HV, as 3.0 × HV and 1.5 × HV, respectively, and 𝑛 is the factor that evaluates the proximity to the mentioned failure criteria of the local stresses at imperfections of the specimens. When the residual stresses are considered, they will be added to the mean applied stress 𝜎 . Table 8 shows the data used for the determination of the 𝑛 factors for both (with residual stresses and without residual stresses for the welded series) conditions and imperfections indicated there. Figure 12 illustrates the position of the fatigue failure points of the specimens according to the mean and alternating local stresses, in relation to respective modified Goodman lines. When examining the concordance between the assumed failure criterion and the cases considered, it was found that it is excellent in the case of the BM, generally good for all cases of the welded series when residual stresses were included and underestimated when they were not included; see Table 8 and Figure 12. There is a good agreement with and without residual stress when the 𝐾 values are low for underfill (cases 2 and 4), and in contrast to the highest values of 𝐾 (cases 1 and 5). Particularly for the excess weld and the underfill of the S2 and S5 series, respectively, Figure 11. Residual stresses profiles in fatigue specimens. 3.5.3. Fatigue Limit Evaluation In previous sections, it was shown that for low stress levels, the approach of reducing fatigue strength through stress-concentrating effect can be used. To consider the effect of the mean stress in the analysis of the fatigue strength at two million cycles of the welded series, due to its simplicity, the modified Goodman line will be used according to the expression: Kfσa Se +Kfσm Sut =1 nG (2) where σa and σm are the nominal alternating and mean stresses applied and both are equal to ∆σ /2 (according to stress ratio R = 0); Kf , is the effective stress concentration factor calculated using the relationship of sensitivity to notch proposed by Peterson [ 37 ]; Sut and Se are the ultimate strength and the fatigue limit of the materials that will be estimated for each welded series and BM through the hardness HV, as 3.0 × HV and 1.5 × HV, respectively, and nG is the factor that evaluates the proximity to the mentioned failure criteria of the local stresses at imperfections of the specimens. When the residual stresses are considered, they will be added to the mean applied stress σm. Table 8shows the data used for the determination of the nG factors for both (with residual stresses and without residual stresses for the welded series) conditions and imperfections indicated there. Figure 12 illustrates the position of the fatigue failure points of the specimens according to the mean and alternating local stresses, in relation to respective modified Goodman lines. When examining the concordance between the assumed failure criterion and the cases considered, it was found that it is excellent in the case of the BM, generally good for all cases of the welded series when residual stresses were included and underestimated when they were not included; see Table 8and Figure 12. There is a good agreement with and without residual stress when the Kf values are low for underfill (cases 2 and 4), and in contrast to the highest values of Kf (cases 1 and 5). Particularly for the excess weld and the underfill of the S2 and S5 series, respectively, there is the greatest discrepancy (close to 20%) between the experimental results and the assumed criterion (cases 3 and 6) when the residual stresses are considered. Metals 2021,11, 1499 17 of 22 Metals 2021, 11, x FOR PEER REVIEW 17 of 22 there is the greatest discrepancy (close to 20%) between the experimental results and the assumed criterion (cases 3 and 6) when the residual stresses are considered. Figure 12. Fatigue failure points of specimens and modified Goodman lines. Table 8. Data for fatigue strength assessment based on the modified Goodman line. Case Condition Series Imperfection K t ρ 𝐾 Hardness Residual Stress Fatigue Limits 𝑛 Factor (mm) (HV) (MPa) (MPa) RS 1 no RS 1 as-welded S1-undercut 3.29 0.06 2.33 335 130 180 1.08 1.60 2 bottom side-removed S1-underfill 1.75 0.53 1.60 335 130 380 0.90 1.10 3 as-welded S2-excess weld 2.25 0.06 1.74 345 90 270 1.20 1.47 4 bottom side-removed S2-underfill 1.77 0.60 1.63 345 90 390 0.94 1.09 5 as-welded S3-undercut 4.41 0.03 2.74 350 50 210 1.05 1.22 6 as-welded S5-underfill 1.95 0.27 1.74 370 200 215 1.17 1.90 7 no-welded Base metal --- --- --- 295 --- 600 --- 0.98 1 RS: residual stresses. The effects of hardness, residual stresses, and stress concentration on each welded series in the as-welded condition can be described as follows: the S1 series presented the lowest fatigue strength due to the combination of a high SCF at the undercut, a medium value of residual stress, and a relatively low hardness. The S5 series, despite having the highest hardness and although the SCF was medium, due to the high value of the residual stresses, it also presented low fatigue strength. The S3 series, with a low value of residual stresses and a high hardness, did not achieve a greater strength to fatigue because it had the highest SCF due to the sharp undercut in the weld root. The S2 series achieved the highest fatigue strength of all series with medium SCF and hardness values and a relatively low residual stress value. It can be inferred that this series can achieve greater fatigue strength if the multiple small imperfections present along the weld root decrease. In other works, the adverse effect of many small imperfections in addition to the weld bead profile on fatigue strength has also been reported [38,39]. Figure 12. Fatigue failure points of specimens and modified Goodman lines. The effects of hardness, residual stresses, and stress concentration on each welded series in the as-welded condition can be described as follows: the S1 series presented the lowest fatigue strength due to the combination of a high SCF at the undercut, a medium value of residual stress, and a relatively low hardness. The S5 series, despite having the highest hardness and although the SCF was medium, due to the high value of the residual stresses, it also presented low fatigue strength. The S3 series, with a low value of residual stresses and a high hardness, did not achieve a greater strength to fatigue because it had the highest SCF due to the sharp undercut in the weld root. The S2 series achieved the highest fatigue strength of all series with medium SCF and hardness values and a relatively low residual stress value. It can be inferred that this series can achieve greater fatigue strength if the multiple small imperfections present along the weld root decrease. In other works, the adverse effect of many small imperfections in addition to the weld bead profile on fatigue strength has also been reported [38,39]. Regarding the fatigue limits achieved by S1 (380 MPa) and S2 series (390 MPa) in the bottom side-removed condition, although the fatigue limit of the S2 series is expected to be higher because its hardness is greater and the residual stresses are lower, the slight difference may be due, as already pointed out in a previous section, to the greater depth of the underfill and to the effect combined with the weld ripples. In the case of the behavior of the S5-F9 specimens, which showed considerably greater fatigue strength than the other specimens of the S5 series, the weld profile in the bottom side was examined with more detail. It was found that the underfill values were below the average of the series and that the variations along the weld axis were small, as illustrated in other work [ 28 ]. The latter implies that the SCF of this series is lower and therefore its fatigue strength is higher. The previous analysis showed that the SCF is a good predictor of fatigue limit in absence of multiple small imperfections, such as those of S2 series that were not easily detected or when the SCF is properly calculated (which was not the case of the S5 series). Conversely, it shows a great effect of the residual stresses that is magnified by the SCF and by the mean applied stress. In summary, although the residual stresses found can be considered small, they adversely affect the fatigue strength of the welded specimens, as well as the high SCFs due to the weld bead geometry and the small associated imperfections. Metals 2021,11, 1499 18 of 22 Meanwhile, the high hardness in the FZ and CG-HAZ allows the improvement of the strength of the welded series. 3.6. Fatigue Limit Predictions The estimates at low stresses showed that the fatigue strength can be predicted if the hardness, residual stresses, weld geometry, and imperfections of the welded joints are known. Therefore, using the expression (2) with nG = 1 and the data in Table 8, the fatigue strength at two million cycles in terms of stress range was predicted and the corresponding results for each welded series and conditions are shown in Table 9together with the experimental results and the relation that compares both results. Considering all the results, there is a good agreement between the experimental and predicted values. However, a difference of greater than 15 percent was found for the S2 and S5 series. For the S2 series, this can be attributed, as already noted, to imperfections in the weld root that the stress concentration factor does not consider. For the S5 series, the difference may be the result of a welding speed range (1.75 to 2.00 m/min) and therefore the value of the SCF, which can actually be higher than the one used ( Kf , 1.95) since the latter was determined based on the average dimensions of the weld beads and at small imperfections in the weld ripples. Table 9. Fatigue limit predictions. Series-Imperfection Condition Fatigue Limits Experimental Predicted Ratio S1-undercut As-welded 180 201 0.90 S1-underfill Bottom side-removed 380 332 1.14 S2-excess weld As-welded 270 336 0.80 S2-underfill Bottom side-removed 390 363 1.07 S3-undercut As-welded 210 222 0.95 S5-underfill As-welded 215 292 0.74 Additionally, regarding the above predictions, and to explain the effect of the imperfections present in the weld root of the S2 series and due to the small size of the undercuts present in the S1 series, the Murakami’s relations proposed to predict the fatigue limit were applied to the mentioned series. The relations have the form: σw=1.43 (HV +120) √Area1/6 1−R 2∝ (3) ∝=0.226 +HV/10,000, (4) where σw (in MPa) is the fatigue limit, HV is the Vickers hardness, R is the stress ratio and √Area (in µ m) is the size parameter of the imperfection. The data shown in Table 10 corresponds in the case of the S1 series to the average size of undercuts, while in the case of the S2 series, it corresponds to the average depth of 10 measurements on the long and narrow imperfections, which are practically continuous in the weld root. In both cases, the √Area parameter was calculated as √Area =a √10 , where ais the crack depth, according to Murakami’s recommendation for this type of imperfection (shallow surface crack) [ 12 ]. In the stress ratio, the residual stresses of Table 8were considered and through an iterative process, in which fatigue limit values were assumed until equality in expression (3) was reached, the fatigue limits were predicted, which are shown in Table 10 as well as the respective comparisons with the experimental results. Metals 2021,11, 1499 19 of 22 Table 10. Data and parameters for prediction of fatigue limits. Series Imperfection Crack Sizes Length, Depth √Area Fatigue Limits (mm, µm) (µm) Experimental (MPa) Predicted (MPa) Ratio S1-undercut 1.11, 40 126 180 214 0.84 S2-shallow crack >0.23, 22 70 270 254 1.06 The results of Table 10 show that adequate predictions can be achieved through the approach proposed by Murakami. In the case of the S2 series, the prediction is better adjusted to the experimental value, while in the case of the S1 series, a similar result was obtained. However, in the predictions based on this last approach, it is necessary to know the imperfection and its size, which may not be evident, as in the case of the S2 series, where only after the fracture of the specimens, they can be observed. Conversely, there may be variations in the sizes of the imperfections, which requires measurements and the application of some criterion that in the present case, was the average size. In the case of undercuts (for series S1), a better prediction can be achieved considering that 4 to 5 undercuts can be present in each specimen; therefore, it is more likely that the size is greater than the average size (40 µ m). Considering that the depth of the undercuts varied from 15 to 105 µ m [ 28 ], for example, an undercut depth of 90 µ m, the fatigue limit will be 180 MPa, which is the same as the experimental result. 4. Conclusions For four laser butt joints in a thin HSLA steel plate, the effects of the weld profile, imperfections, hardness and residual stresses were considered to explain the experimental S-N curves and for predictions at low stress levels, applying approaches based on both stress-concentrating effect and Murakami’s relationship. The main conclusions drawn are: • The fatigue strength of the welded series exceeded the FAT100 reference curve with fatigue limits in the range of 180 to 340 MPa, presenting multiple imperfections as shallow but sharp, and considered as crack-like imperfections and with B and D quality levels according to the ISO13919-1 standard; • Although each series presented several fatigue failure modes due to the presence of different imperfections in the weld bead, a dominant failure mode strongly related to the imperfection with the highest SCF was observed for each series: 3.29—undercut—S1, 2.25—excess weld—S2; 4.41—undercut—S3; 1.95—underfill—S5; • At low stress levels, the local properties of each welded series: hardness, residual stress, and weld bead geometry, determined the fatigue limits and their trends were in good agreement with the experimental results. Both high SCF and relatively low residual tensile stresses (<200 MPa) affected the fatigue limits; only when SCFs are low (<2.0) can the level (<130 MPa) of residual tensile stresses not be considered, while the local hardness increased the fatigue limits of the welded series although its influence is lower than that of the other two factors; • The predictions of the fatigue limits were acceptable by the stress-concentrating effect when there were no small imperfections associated with excess weld, undercut, or underfill, while when these were present or for small undercuts, the relationship proposed by Murakami was appropriate. Author Contributions: Conceptualization, P.G.R. and J.F.; methodology, P.G.R., A.C.B., C.C., and J.F.; software, P.G.R.; validation, P.G.R., F.A., and A.C.B.; formal analysis, P.G.R.; investigation, P.G.R.; writing—original draft preparation, P.G.R.; writing—review and editing, F.A., A.C.B., and C.C.; supervision, J.F. All authors have read and agreed to the published version of the manuscript. Metals 2021,11, 1499 20 of 22 Funding: The authors acknowledge the support provided by the Universidad de las Fuerzas Armadas-ESPE for the scholarship and Mário da Costa Martins and Fernando Meireles of the Orthopedia Medica Company for laser welding manufacturing. This research is sponsored by FEDER funds through the program COMPETE–Programa Operacional Factores de Competitividade— and by national funds through FCT–Fundação para a Ciência e a Tecnologia–, under the project UIDB/00285/2020. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conflicts of Interest: The authors declare no conflict of interest. Nomenclature aCrack depth √Area Murakami’s area parameter Cparameter of the S-N curve HV Vickers hardness mSlope of S-N curve in log-log scales NNumber of cycles KtStress concentration factor KfEffective stress concentration factor RStress ratio equal to minimum stress divided by maximum stress nGGoodman criteria factor Sut Tensile ultimate strength SeFatigue limit ∆σNominal stress range $Imperfection radius σaNominal alternating stress σmNominal mean stress σwFatigue limit according Murakami’s expression ∝Murakami’s factor Abbreviations ASTM American Society for Testing and Materials BM Base metal CG-HAZ Corse-grain heat-affected zone FAT Fatigue class curve FEM Finite element method FG-HAZ Fine-grain heat-affected zone FCG Fatigue crack growth FCGR Fatigue crack growth rate FM Failure mode FZ Fusion zone HAZ Heat-affected zone HSLA High-strength low-alloy HV Hardness Vickers IIW International Institute of Welding ISO International Organization for Standardization NDT Nondestructive testing HI Heat input SCF Stress concentration factor References 1. Guo, W.; Wan, Z.; Peng, P.; Jia, Q.; Zou, G.; Peng, Y. Microstructure and mechanical properties of fiber laser welded QP980 steel. J. Mater. Process. Technol. 2018,256, 229–238. [CrossRef] 2. Farabi, N.; Chen, D.; Zhou, Y. Fatigue properties of laser welded dual-phase steel joints. Procedia Eng. 2010 ,2, 835–843. [CrossRef] Metals 2021,11, 1499 21 of 22 3. Xu, W.; Westerbaan, D.; Nayak, S.; Chen, D.; Goodwin, F.; Zhou, Y. Tensile and fatigue properties of fiber laser welded high strength low alloy. Mater. Des. 2013,43, 373–383. [CrossRef] 4. Gurney, T. Cumulative Damage of Welded Joints; Woodhead Publishing Ltd.: Cambridge, UK, 2006. 5. Hobbacher, A.F. Recommendations for Fatigue Design of Welded Joints and Components; Springer: London, UK, 2016. 6. Maddox, S.J. Fatigue Strength of Welded Structures; Abington Publishing: Cambridge, UK, 2002. 7. Chapetti, M.; Jaureguizahar, L. Fatigue behavior prediction of welded joints by using an integrated fracture mechanics approach. Int. J. Fatigue 2012,43, 43–53. [CrossRef] 8. Zhang, Y.-H.; Maddox, S. Fatigue life prediction for toe ground welded joints. Int. J. Fatigue 2009,31, 1124–1136. [CrossRef] 9. Yamamoto, H.; Danno, Y.; Ito, K.; Mikami, Y.; Fujji, H. Weld toe modification using spherical-tip WC tool FSP in fatigue strength improvement of high-strength low-alloy steel joints. Mater. Des. 2018,160, 1019–1028. [CrossRef] 10. Mei, J.; Xiang, S.; Vasu, A.; Chung, J.; Desai, R.; Dong, P. The fatigue limit prediction of notched components—A critical review and modified stress gradient based approach. Int. J. Fatigue 2020,135, 105531. [CrossRef] 11. Schijve, J. The significance of fatigue crack initiation for predictions of the fatigue limit of specimens and structures. Int. J. Fatigue 2014,61, 39–45. [CrossRef] 12. Murakami, Y. Metal Fatigue: Effects of Small Defects and Nonmetallic Inclusions; Elsevier Science Ltd.: Oxford, UK, 2002. 13. Åman, M.; Tanaka, Y.; Murakami, Y.; Remes, H.; Marquis, G. Fatigue strength evaluation of small defect at stress concentration. Struct. Integr. Procedia 2017,7, 351–358. [CrossRef] 14. Lassen, T.; Récho, N. Fatigue Life Analyses of Welded Structures; ISTE Ltda: London, UK, 2006. 15. Lieurade, H.P.; Huther, I.; Lefebvre, F. Effect of Weld Quality and Postweld Improvement Techniques on the Fatigue Resistance of Extra High Strength Steels. Weld. World 2008,52, 106–115. [CrossRef] 16. Gerritsen, C.; Vanrostenberghe, S.; Doré, M. Diode laser weld toe re-melting as a means of fatigue strength improvement in high strength steels. Procedia Eng. 2013,66, 171–180. [CrossRef] 17. Cheng, X.; Fisher, J.; Prask, H.; Yen, B.; Graunapel-Herold, T.; Roy, S. Residual stress modification by post-weld treatment and its beneficial effect on fatigue strength of welded structures. Int. J. Fatigue 2003,25, 1259–1269. [CrossRef] 18. Lefebvre, F.; Peyrac, C.; Elbel, G.; Revilla-Gomez, C.; Verdu, C.; Buffière, J. Understanding of fatigue strength improvement of steel structures by hammer peening treatment. Procedia Eng. 2015,133, 454–464. [CrossRef] 19. Marquis, G.; Barsoum, Z. A guideline for fatigue strength improvement of high strength steel welded structures using high frequency mechanical impact treatment. Procedia Eng. 2013,66, 98–107. [CrossRef] 20. Maddox, S. Fatigue Design Rules for Welded Structures. In Fracture and Fatigue of Welded Joints and Structures; Woodhead Publishing: Cambridge, UK, 2011; pp. 168–207. 21. Lillemäe, I.; Remes, H.; Liinalampi, S.; Itävuo, A. Influence of weld quality on the fatigue strength of thin normal and high strength steel butt joints. Weld. Word 2016,60, 731–740. [CrossRef] 22. Kucharczyk, P.; Madia, M.; Zerbst, U.; Schork, B.; Gerwien, P.; Münstermann, S. Fracture-mechanics based prediction of the fatigue strength of weldments. Material aspects. Eng. Fract. Mech. 2018,198, 79–102. [CrossRef] 23. Li, S.; Khan, Y.; Kuang, S. Effects of microstructure on fatigue crack growth behavior in cold-rolled dual phase steels. Mater. Sci. Eng. A 2014,612, 153–161. [CrossRef] 24. Watanabe, O.; Matsumoto, S.; Nakano, Y.; Saito, Y. Fatigue strength of welded joints in high strength steel Effects of stress concentration factor and welding residual stress. Weld. Int. 1996,10, 201–206. [CrossRef] 25. Harati, E.; Karlsson, L.; Svensson, L.-E.; Dalaei, K. The relative effects of residual stresses and weld toe geometry on fatigue life of weldments. Int. J. Fatigue 2015,77, 160–165. [CrossRef] 26. Nguyen, N.; Wahab, M. The Effect of Undercut and Residual Stresses on Fatigue Behaviour of Misaligned Butt Joints. Eng. Fract. Mech. 1996,55, 453–469. [CrossRef] 27. Shiozaki, T.; Yamaguchi, N.; Tamai, Y.; Hiramoto, J.; Ogawa, K. Effect of weld toe geometry on fatigue life of lap fillet welded ultra-high strength steels joints. Int. J. Fatigue 2018,116, 409–420. [CrossRef] 28. Riofrío, P.; Capela, C.; Ferreira, J. Imperfections and Modelling of the Weld Bead Profile of Laser Butt Joints in HSLA Steel Thin Plate. Metals 2021,11, 151. [CrossRef] 29. ISO. Welding—Electron and Laser-Beam Welded Joints—Guidance on Quality Levels for Imperfections. Part 1: Steel; ISO 13919-1(1996); ISO: Geneva, Switzerland, 1996. 30. Data Sheet S700MC. 2017. Available online: http://www.ssab.com (accessed on 1 February 2019). 31. Riofrío, P.; Capela, C.; Ferreira, J.; Ramalho, A. Interactions of the process parameters and mechanical properties of laser butt welds in thin high strength low alloy steel plates. J. Mater. Des. Appl. 2020,234, 665–680. [CrossRef] 32. ASTM International. Standard Practice for Conducting Force Controlled Constant Amplitude Axial Fatigue Tests of Metallic Materials; E466-15; ASTM International: West Conshohocken, PA, USA, 2015. 33. ASTM International. Standard Practice for Statistical Analysis of Linear or Linearized Stress-Life (S-N) and Strain-Life ( ε -N) Fatigue Data; E739-10; ASTM International: West Conshohocken, PA, USA, 2015. 34. Remes, H.; Varsta, P. Statistics of weld geometry for laser-hybrid welded joints and its application within notch stress approach. Weld. World 2010,54, R189–R207. [CrossRef] 35. Fricke, W. IIW Recommendations for the Fatigue Assessment of Welded Structures by Notch Stress Analysis; Woodhead Publishing Limited: Oxford, UK, 2012. Metals 2021,11, 1499 22 of 22 36. Alam, M.; Barsoum, Z.; Jonsén, P.; Kaplan, A.; Haggblad, H. The influence of surface geometry and topography on the fatigue cracking behaviour of laser hybrid welded eccentric fillet joints. Appl. Surf. Sci. 2010,256, 1936–1945. [CrossRef] 37. Pilkey, W. Peterson’s Stress Concentration Factors; John Wiley & Sons: New York, NY, USA, 1997. 38. Schork, B.; Zersbt, U.; Kiyak, Y.; Kaffenberger, M.; Madia, M.; Oechsner, M. Effect of the parameters of weld toe geometry on the FAT class as obtained by means of fracture mechanics-based simulations. Weld. Word 2020,64, 925–936. [CrossRef] 39. Otegui, J.; Kerr, H.; Burns, D.; Mohaupt, U. Fatigue Crack Initiation from Defects at Weld Toes in Steel. Int. J. Press. Vessel. Pip. 1989,38, 385–417. [CrossRef]