Vol.:(0123456789)1 3 Welding in the World (2023) 67:2791–2802 https://doi.org/10.1007/s40194-023-01600-9 RESEARCH PAPER Mechanical properties of MAG butt welded dissimilar structural steel joints with varying strength from grade S355 up to S960 Oliver Brätz1  · Mareike von Arnim2 · Stefan Eichler3 · Andreas Gericke1 · Jörg Hildebrand3 · Jean Pierre Bergmann3 · Ulrike Kuhlmann2 · Knuth‑Michael Henkel1,4 Received: 25 July 2023 / Accepted: 28 September 2023 / Published online: 14 October 2023 © The Author(s) 2023 Abstract Mixed connections made of normal-strength and high-strength structural steels allow for optimized material usage and production effort in applications where, as a result of different mechanical effects on materials of the same type, it would otherwise be necessary to adjust the plate thickness. Reduced material consumption and smaller weld geometries can thus generate ecological and economic advantages. When welding high-strength structural steels, however, significant soften- ing can occur in the heat-affected zone, which can influence the load-carrying behavior of the overall joint. Since there are currently no appropriate standards for butt welds made of steels with different strengths up to S960, a separate design concept is required. In this paper, the weldability and load-carrying capacity of multilayer MAG welded butt joints designed as mixed connections of a normal-strength structural steel S355 and a high-strength structural steel in the range S690 to S960 are investigated. Extensive experimental investigations are carried out, in which other influencing variables such as the filler metal used, the heat input, the plate thickness, and the weld geometry are varied in order to identify their effects on the load-carrying capacity of the welded joints. Among other things, the results form the basis for an empirically based design model for mixed connections. Keywords High-strength steel · Mixed connections · Joint design · Load-carrying capacity · Weldability · Impact toughness 1 Introduction The use of high-performance weldable construction steels with increased strength allows the design of slender and filigree structures with long spans (see Fig. 1a). The high strength-to-weight ratio of these materials results in enormous advantages regarding the design weight, and therefore induce fewer material resource consumption and cost savings. Especially in the tensile area, cross-sec- tional dimensions can be reduced. As a secondary effect, the material requirements for the substructures can also be reduced, resulting in an improved life cycle assess- ment of the entire structure. In order to use structural steels with higher strength, commonly categorized according to their nominal yield strength as high-strength steels (HSS: fy = 460–690 MPa) and ultra-high strength steels (UHSS: fy ≥ 690–1100 MPa), efficiently in structures, there may be a need for welded butt joints being mixed connections of normal-strength steel (fy = 355 MPa) and HSS or UHSS. In bridge construction, cranes, offshore structures, or wind tur- bines, for example, the material load capacity can thus be adapted to the moment curve without having to adjust the plate thicknesses of the joint partner (cf. [1]). As a result, increased fabrication costs in the execution of thickness vari- ations can be reduced, such as the mechanical preparation of the plates in the area of butt joints (see Fig. 1b). In addition, the reduced plate thicknesses result in a smaller weld seam Recommended for publication by Commission XV - Design, Analysis, and Fabrication of Welded Structures * Oliver Brätz oliver.braetz@igp.fraunhofer.de 1 Fraunhofer Institute for Large Structures in Production Engineering IGP, Albert-Einstein-Straße 30, 18059 Rostock, Germany 2 Institute of Structural Design, University of Stuttgart, Pfaffenwaldring 7, 70569 Stuttgart, Germany 3 Production Technology, Department of Mechanical Engineering, Technische Universität Ilmenau, Gustav-Kirchhoff-Platz 2, 98693 Ilmenau, Germany 4 Chair of Joining Technology, University of Rostock, Albert-Einstein-Straße 30, 18059 Rostock, Germany http://orcid.org/0000-0002-7539-963X http://crossmark.crossref.org/dialog/?doi=10.1007/s40194-023-01600-9&domain=pdf 2792 Welding in the World (2023) 67:2791–2802 1 3 volume, so that enormous potential cost savings in produc- tion result from reduced welding times, filler materials, and possibly reduced preheating efforts. Notwithstanding, for certain fields, the lack of relevant experimental tests and limitations of the existing design standards due to uncer- tainty about the behavior of HSS/UHSS still hinder their wider applications in engineering practice [2–4]. Typical issues with regard to the metallurgical weldability of HSS/ UHSS are, e.g., cold cracking in the heat-affected zone (HAZ) or weld metal, HAZ softening, hot cracking in weld metal, deterioration of toughness, and lack of ductility in HAZ [5–10]. The use of butt welds as mixed connections of steels with dissimilar strength is currently not clearly regulated by the European design code EN 1993-1-8 [11] for joint design in steel structures. Therefore, a research program funded by the German Federal Ministry for Economic Affairs and Climate Action was initiated in 2021, and comprehensive experi- mental as well as numerical studies were performed, see Acknowledgements. Different plate thicknesses (t = 3, 10 and 20 mm), different groove types, different high-strength steel grades up to S960, different welding consumables (solid and metal cored wires with different filler metal strength), and heat inputs were used. With the results of the experimental and numerical investigations as well as further accompany- ing studies on the material properties, design rules for butt welds as mixed connections are being developed, which will in future influence the design rules in the second-generation of Eurocodes. In this paper, the weldability, i.e., the metal- lurgical behavior in particular, and the experimental results of investigation on the load-carrying capacity for mixed con- nections of construction steels with dissimilar strength and plate thicknesses of t = 10 and 20 mm are presented. The different welding behavior of dissimilar joined steels may cause a challenging metallurgy. For welding with filler metal, part of investigation was to predict the structure of the fusion zone and the risks associated with the different metal compositions. Moreover, welding of UHSS often involves consumables with lower strength, viz., undermatching filler materials [4, 12]. On one side, the selected welding parameters, i.e., heat input, effect the resulting weld metal microstructure, and thus need to be in a specific range to ensure certain material properties. On the other side, weld- ing related HAZ formations differ for the dissimilar joint partner based on their composition and steel making process route and therefore must be considered, too [8]. For weld- ing of S960MC with undermatching G46 filler metal, it was found that the resulting material properties differ significantly depending on the used welding process. In contrast to con- ventional gas metal arc welding (GMAW), joints realized by plasma welding reached the strength and toughness require- ments of the base material. For GMAW, the use of a matching filler is necessary to achieve sufficient strength [12]. In con- trast to typical normalized (N) low alloyed structural steels, when welding high-strength structural steels in quenched and tempered condition (QT) or as thermo-mechanically con- trolled process (TMCP or TM) steel, softening may occur within the heat-affected zone (HAZ) due to transformations and tempering effects, depending on the composition and the manufacturing process [13–15]. Depending on the extent of this soft zone with locally reduced strength, the load-carrying capacity of the entire joint can be negatively affected [6, 7, 16]. In some cases, a reduction of the load-carrying capaci- ties by more than 15% has been observed compared to the ultimate tensile strength of the UHSS [16, 17]. Currently, no separate design checks are usually required for butt welds if the filler metal reaches at least the yield strength and ultimate tensile strength of the base material [11]. Accordingly, the general member check for the base material acc. to EN 1993- 1-1 [18] is sufficient. The current European standards for the design of steel structures (Eurocode 3) for joints EN 1993-1-8 [11, 19] and EN 1993–1–12 [20, 21] as extension up to steel grades S700 are insufficient for higher strength butt welds because the design is partly on the unsafe side [17]. These standards do not take into account the special metallurgical weldability of HSS/UHSS. For example, the formation of a soft zone in the HAZ is not covered. On the basis of experimental Fig. 1 a Application of high-strength steels, and b butt welds with individual load on joint partner executed with unequal thickness and chamfer- ing of the thicker part vs. mixed connection of dissimilar steel strength with equal thickness 2793Welding in the World (2023) 67:2791–2802 1 3 and numerical investigations [16, 22], a design concept has already been developed which allows an economical and safe design of high-strength butt welds. The concept applies to full penetration butt welds of steel grades greater than S460 and up to S700 using filler material with lower ('under- matching'), equal ('matching'), and higher ('overmatching') strength compared to the base material. The design concept has already been accepted by the responsible working group CEN TC250/SC 3/WG 8 for the final draft European stand- ard FprEN 1993-1-8 [23]. Since the new edition prEN 1993-1-12 [24] is now to be extended for steel grades up to S960, tests were carried out on butt welds of similar steel grades S960QL as representa- tive of a QT steel or S960MC as a TMCP steel [2]. For weld- ing with matching filler materials, it was found that the new design model from FprEN 1993-1-8 leads to good agree- ments. However, for joints made with significantly under- matching G46 filler material, an adjustment of the model is necessary. With the results obtained, the design equation will then be extended up to S960. On welded joints of HSS up to S890 with unequal weld metal strength ('mismatch'), a significant influence on the failure behavior of the joint as a function of the mismatch ratio was found [17]. For mixed joints of normal-strength and high-strength steel up to S700, it is currently valid acc. to EN 1993-1-8 [11, 19] and EN 1993-1-12 [20, 21] for full penetration butt welds that the load-carrying capacity is equivalent to that of the weaker connected component. However, this has not yet been demonstrated for UHSS of the grade S960. In addition, the filler metal’s specified strength (Rp0.2 and Rm), elongation at failure and Charpy V–notch (CVN) energy value should be equivalent to or better than that specified for the parent material [11]. Consequently, the filler metal must be oriented to the higher-strength steel. In this research project, see funding reference, extensive experimental and numerical investigations were carried out to analyze whether this procedure is necessary and whether the formation of a soft zone in mixed joints is relevant for the load-carrying and deformation capacity of the butt welds. 2 Experimental methodology 2.1 Material and sample welding Various structural steel grades with different process routes (N, QT, and TMCP) and strengths were used. They cor- respond to the specified European standard technical deliv- ery conditions for hot-rolled products made of structural steels (see Table 1). Both solid wire electrodes and metal cored wire electrode were used as consumables for metal active gas (MAG) welding using an Ar/CO2 gas mixture (ISO 14175-M21-ArC-82/18) 15 l/min as shielding gas. The compositions of the used materials and the carbon equiva- lents representing their metallurgical weldability regarding welding-related hardening are listed in Table 2. The weld samples were made fully mechanized by using a robot (Fig. 2a). Butt welds with a length of 600 mm were welded in flat position (ISO 6947-PA/1G acc. AWS). The width of the plates was 300 mm each (Fig. 2b). The roll- ing direction of the plates was oriented longitudinally to the weld. Either a single-V groove with an included angle of 60° and a 2-mm root face was one-side welded, or a 60° double- V groove with a 2-mm root face was double-side welded. The parameters for weld sample production are listed in Table 3 as average values obtained for the multi-pass welds. To investigate the influence of the heat input due to welding, different arc energies were selected. The 'low' arc energy was Table 1 Used base materials and filler metals, specifications, and actual material properties Material Specification Dimension ReH or Rp0.2 in MPa Rm in MPa A in % KV2 in J (at T) S355J2+N, t10 EN 10025-2-S355J2+N (1.0577) t = 10 mm 477 571 34 130 (−20 °C) S355J2+N, t20 t = 20 mm 400 541 31 153 (−20 °C) S690QL, t10 EN 10025-6-S690QL (1.8928) t = 10 mm 859 887 14 165 (−40 °C) S690QL, t20 t = 20 mm 851 883 17 170 (−40 °C) S700MC, t10 EN 10149-2-S700MC (1.8974) t = 10 mm 792 823 19 121 (−20 °C) S960QL, t10 EN 10025-6-S960QL (1.8933) t = 10 mm 1036 1077 11 27 (−40 °C) S960QL, t20 t = 20 mm 1045 1074 14 138 (−40 °C) G46 ISO 14341-A G 46 5 M21 4Si1 Ø1.2 mm 525 598 22 83 (−50 °C) G69 ISO 16834-A G 69 6 M21 Mn4Ni1.5CrMo Ø1.2 mm 761 806 18 81 (−60 °C) G79 ISO 16834-A G 79 5 Mn4Ni1.5CrMo Ø1.2 mm 889 940 17 57 (−50 °C) G89 ISO 16834-A G 89 6 M21 Mn4Ni2CrMo Ø1.2 mm 1040 1108 15 62 (−60 °C) T69 ISO 18276-A T 69 6 Mn2NiMo M M21 2 H5 Ø1.2 mm 755 790 20 80 (−60 °C) T89 ISO 18276-A T 89 4 Z M M21 3 H5 Ø1.2 mm 931 993 19 82 (−40 °C) 2794 Welding in the World (2023) 67:2791–2802 1 3 defined as welding processes that lead to a relatively short target cooling time of t8/5 = 5 s, describing the time interval for the cooling span from 800 to 500 °C of the weld metal. Corresponding to the high-strength steel involved, the 'high' arc energy corresponds to a defined target cooling time of t8/5 = 8 or 12 s, which represent the maximum processing recommendations specified by the steel manufacturer. The required lowest preheating temperatures were determined according to common concepts for the avoidance of hydro- gen-assisted cold cracking based on the carbon equivalent CET of the filler metals and base metals, the geometry (form of weld and plate thickness), the heat input, and the expected maximum diffusible hydrogen contents in weld metal acc. to ISO/TR 17671-2 [25]. The present cooling times t8/5 were determined by means of an infrared camera on the solidified weld beads (cf. [10]) (see Fig. 2a). Due to the different arc energies, the number of weld beads varied for the different samples to ensure full penetration joints. The samples were cooled in air or preheated again to ensure interpass tempera- tures equal to the preheating temperatures. The test program comprises 36 weld samples with t = 10 mm and 24 weld samples with t = 20 mm. Three tensile test pieces were extracted per weld sample. Various combinations of base metal pairing (in each case normal strength S355J2 + N in combination with a high yield strength structural steel of grade S690QL/S700MC/S960QL), filler material (G46 as standard for S355 or a solid wire/tubular metal cored wire as matching filler metal with respect to the HSS/UHSS), heat input (high and low), and weld geometry (each as single-V and double-V groove) were investigated to cover a relevant range of applications. 2.2 Weld formation and hardness distribution To investigate the metallurgical weldability and the local material condition, the welded joints were analyzed in trans- verse cross sections. The specimens were metallographically prepared and etched with alcoholic nitric acid, 3% Nital. Subsequently, microhardness mapping with 50-µm-grid was determined using an automatic tester (BAQ, UT 200) based on the ultrasonic contact impedance (UCI) method acc. to Table 2 Chemical composition of the used base materials and filler metals 1 Carbon equivalent acc. ISO/TR 17671-2 – Method B: CET = C + (Mn + Mo)/10 + (Cr + Cu)/20 + Ni/40 in % Material C wt.% Si wt.% Mn wt.% Cr wt.% Ni wt.% Mo wt.% Cu wt.% P + S ppm Al ppm Ti ppm B ppm Nb ppm V ppm N ppm CET1) in % S355J2+N, t10 0.150 0.21 1.53 0.06 0.03 0.01 0.02 170 360 30 n.a 200 0 40 0.31 S355J2+N, t20 0.150 0.18 1.52 0.04 0.03 0.01 0.02 200 450 20 n.a 200 0 50 0.31 S690QL, t10 0.140 0.21 1.03 0.33 0.05 0.22 0.01 140 920 60 30 270 0 40 0.28 S690QL, t20 0.140 0.23 1.17 0.34 0.05 0.31 0.03 121 820 40 28 290 20 30 0.31 S700MC, t10 0.063 0.03 1.86 0.03 0.01 0.01 0.01 97 620 1350 2 530 70 n.a 0.25 S960QL, t10 0.167 0.29 1.39 0.28 0.12 0.52 0.10 100 270 20 28 260 20 50 0.38 S960QL, t20 0.165 0.28 1.08 0.19 0.04 0.52 0.02 120 n.a 180 14 10 10 n.a 0.34 G46 0.080 0.96 1.67 0.03 0.01 0.01 0.01 180 20 110 4 10 20 n.a 0.25 G69 0.080 0.56 1.65 0.24 1.58 0.47 0.03 170 30 640 10 40 30 n.a 0.34 G79 0.100 0.76 1.74 0.32 1.82 0.53 0.03 190 50 58 10 30 40 n.a 0.39 G89 0.110 0.83 1.78 0.37 2.24 0.57 0.01 170 30 750 1 30 30 n.a 0.42 T69 0.050 0.43 1.64 0.02 2.30 0.55 0.01 180 n.a n.a n.a n.a 10 n.a 0.33 T89 0.090 0.51 1.29 0.55 2.80 0.82 n.a 220 n.a n.a n.a n.a n.a n.a 0.40 Fig. 2 Welding of specimens: a weld setup incl. cooling time determination, and b weld sample geometry and tensile test piece position (dimen- sion in mm) 2795Welding in the World (2023) 67:2791–2802 1 3 DIN 50159-1 [26]. A conversion of the results into the Vick- ers scale in HV 0.1 (UCI) was carried out. Validation of the UCI hardness measurement was previously made using reference measurements made with a conventional Vickers hardness tester (Struers, Durascan) (cf. [27]). 2.3 Mechanical weld testing To investigate the load-carrying capacity, quasi-static trans- verse tensile tests until specimen break were carried out. Different test piece geometries were defined according to the plate thicknesses (see Fig. 3a). Three test specimens were taken from each weld samples in the central area of the weld length (see Fig. 2b). The weld reinforcement/ convexity was removed by grinding. The tensile tests were performed on a 1-MN-testing machine with constant dis- placement-controlled testing speed. Specimens with plate thickness t = 10 mm were tested at 2 mm/min and specimens with plate thickness t = 20 mm were tested at 3 mm/min. For a more detailed analysis of the deformation behavior under loading, a spatially resolved strain measurement was performed using a high-resolution 3D camera system for Table 3 Mean welding parameter for sample manufacturing 1)  Target cooling time given in brackets as t8/5 in s 2)  Arc type/transfer mode: short circuit (sc), globular (g), spray (s) or pulse (p) 3)  Mean value for standard arc welding, RMS value for pulse welding 4)  Arc energy as heat input without consideration of an efficiency factor acc. to ISO/TR 18491 Filler metal Thick- ness t in mm Joined HSS to S355J2+N Heat input1) Preheating TP in °C Arc type2) Current I in A3) Voltage U in V3) Arc energy E in kJ/ mm4) Cooling time t8/5 in s Solid wire: G46, G69, G79 or G89 10 S690QL Low (5) 25 p 214 25.0 0.82 5.7 S700MC, S960QL 120 sc 207 22.6 0.61 5.2 S960QL High (8) 120 g 255 24.7 0.79 7.7 S690QL High (12) 25 s 307 28.2 1.21 11.7 S700MC 120 s 272 28.3 0.97 13.0 20 S690QL Low (5) 170 p 208 25.3 0.81 4.3 S960QL 150 g 247 25.0 0.83 5.2 S690QL High (8) 170 s 304 28.3 1.20 7.3 S960QL 150 s 311 27.3 1.13 8.6 Metal cored wire: T89 or T69 10 S690QL, Low (5) 25 s 298 25.9 1.03 6.6 S960QL 120 g 187 29.8 0.74 5.4 S960QL High (8) 120 s 218 29.2 0.85 8.3 S690QL High (12) 25 s 325 26.7 1.19 9.9 20 S690QL Low (5) 170 s 290 25.9 1.00 5.4 S960QL 150 g 232 29.4 0.91 5.1 S690QL High (8) 170 s 321 26.8 1.17 7.8 S960QL 150 s 312 29.1 1.21 8.7 Fig. 3 Transverse tensile testing of the butt weld: a test piece geometry, and b test setup with optical strain measurement 2796 Welding in the World (2023) 67:2791–2802 1 3 full-field measurements (GOM, ARAMIS). Here, the dis- tortion of a previously applied, irregular gray-scale pattern was recorded and evaluated (see Fig. 3b) with exemplary testing of a specimen with the same material pairing made of S960 [2]. In addition, circular cross-section test pieces (Ø10 mm, original gauge length L0 = 50  mm) were taken from selected single-V weld samples t = 20 mm to characterize the strength and ductility of the weld metal. Charpy impact testing acc. to ISO 148–1 with standard V-notched test pieces (10 × 10 × 55  mm3) transverse to weld was carried out to investigate the toughness properties in the welded joint. Since the direct comparison of CVN results is only of significance when made between test pieces of the same dimensions, only welds from t = 20 mm samples were tested to ensure extraction of standard size test pieces. The testing was performed by using a 450-J-pendulum (PSd 450, WPM Leipzig) equipped with a 2-mm striker. In accordance with a welding procedure qualification acc. to ISO 15614-1 [28], a set of three test pieces in the weld metal and in the HAZ each were tested. The specified impact properties required by base metal’s technical delivery conditions must be fulfilled for the average value. One individual value may be below the speci- fied value, provided that it is not less than 70% of that value. For joints of dissimilar metal, impact tests shall be carried out in the HAZ of both parent metal and the absorbed energy shall be in accordance with the parent material standard. For the weld metal, the specimen location and notch orientation correspond to VWT 0/1 acc. to ISO 9016 [29], i.e., the notched face through plate thickness and position in the center line of the weld metal. The distance from the weld joint face side to the nearer face of the test specimen was 1 mm. For impact tests of the HAZ, specimens were taken with same notch geometry and at a 1-mm distance of the notch mid-point from the reference line, i.e., fusion line (VHT 1/1). The test temperatures were selected according to the specified impact properties of the base material with higher requirements. 3 Results and discussion 3.1 Weldability For the assessment of weldability acc. to ISO/TR 581 [30], different aspects are to be evaluated together, i.e., metallurgical weldability as a material property, con- structional weldability in virtue of its design, and opera- tive weldability with respect to production conditions. The operative weldability of the examined processes is generally considered to be given due to the execution of common welding tasks (incl. joint preparation, preheating and execution of the welding work). The constructional weldability of the component under the intended operat- ing loads and conditions must be taken into account in the design. The metallurgical weldability of the base materi- als and filler metals used was investigated for the welding procedures with different material pairings, heat inputs, and joint geometries regarding weld metallurgy, strength, and impact energy. For all applied procedures, no metallurgical problems such as hot or cold cracking, or excessive embrittlement was found despite the sometimes very different base metal/ filler metal combinations and the resulting weld metal com- position (cf. [4]). The mechanical properties analyzed were strength by means of weld metal tensile tests and hardness distribution analysis in sectioned joint, and toughness by means of notched bar impact testing. The hardness mapping across the joint area shows the transformation behavior and reveals the resulting inhomo- geneous strength. The average hardness of the weld metal was found at a higher level for lower arc energies as a result of the accelerated cooling and the associated increasingly displacive phase transformation into microstructural con- stituents having high hardness. This is particularly evident in the cover pass region with slight reheating interaction, compare Fig. 4a and b, as well as e and f. The average hardness of the G69 weld metal for the sin- gle-V groove (t = 20 mm) was about 340 HV 0.1 (UCI) for the lower arc energy of E = 0.8 kJ/mm (t8/5 ≈ 4 s) and about 315 HV 0.1 (UCI) for E = 1.19 kJ/mm (t8/5 ≈ 7.5 s). For the slower cooled G79 weld metal, there was an average hardness of about 310 HV 0.1 (UCI) for E = 0.85 kJ/mm (t8/5 ≈ 12 s) and about 290 HV 0.1 (UCI) for E = 1.1 kJ/ mm (t8/5 ≈ 18 s). The comparison of the hardness of G89 weld metal of a single-V and double-V-groove weld does not reflect the influence of an increased proportion of reheated microstructure or the lower degree of dilution. At compara- ble heat input, this resulted in about 375 HV 0.1 (UCI) for the double-V-groove weld (t8/5 ≈ 8.4 s) and 390 HV 0.1 (UCI) for the single-V-groove weld (t8/5 ≈ 7.5 s). At the side of S355J2 + N, a typical formation of the HAZ with characteristic regions depending on the temperature cycle was found, i.e., a considerably hardened area of up to about 350 HV 0.1 (UCI) in the coarse-grained region close to the fusion line (CGHAZ, Tmax ≫ Ac3), a fine-grained/ normalized region (FGHAZ, Tmax > Ac3) that was slightly influenced in terms of strength, as well as the intercriti- cally heated region (ICHAZ, Ac1 < Tmax ≤ Ac3) and subcriti- cal heated region (SCHAZ, Tmax < Ac1) that were slightly affected. With increased energy per unit length, the HAZ increased, but due to the associated longer cooling times, a microstructure with a tendency to a lower hardness level was formed, especially in the CGHAZ (cf. Figure 4a and b) on the right. The strength level of the normal-strength steel was not negatively affected by welding. 2797Welding in the World (2023) 67:2791–2802 1 3 The high-strength QT steels S690QL and S960QL exhib- ited a hardened region in the CGHAZ and FGHAZ in a range of approximately 410–430 HV 0.1 (UCI). In the ICHAZ and SCHAZ, a lower hardness level of approximately 250–290 HV 0.1 (UCI) was present which was below the hardness of the unaltered base material being about 310 and 360 HV 0.1 (UCI), respectively. Especially in the SCHAZ, tempering effects led to significant softening (see Fig. 4a to d left) [5, 15]. As a result of the multilayer welding proce- dure with locally superimposed temperature–time cycles, a complex microstructure resulted. The subcritically and inter- critically reheated areas of the previous weld bead’s HAZ were softened, too. Comparing the single-V and double-V- groove weld, a wider soft zone was present due to one-sided welding. In the case of welding on both sides, there was a larger proportion of the cover pass’ HAZ over the plate thickness with as welded microstructure without reheating, so that an overall higher strength level was obtained in the joint, compare Fig. 4c and d, left. Based on the determined hardness mapping, it can be assumed that the permitted maximum hardness values within the scope of a procedure qualification acc. ISO 15614–1 [28] of 380 and 450 HV10 for N/TMCP steels and QT steels, respectively, can be met. The weld metal strength was influenced by the different heat inputs or the associated resulting cooling times as well as the layer structure (see Fig. 5a). The G46 weld metal, which was welded with high arc energy containing a lower number of weld passes, exhibited higher strength. On the other hand, for the weld metal made of high-strength filler material G89 or T89, it was evident that a lower strength level was obtained with increasing cooling time. Only the G46 weld metal displayed a pronounced yield point in the stress–strain curve. The ductility of the high-strength weld metal was significantly lower than that of the G46. Moreo- ver, high yield strength ratios of Rp0.2/Rm > 0.92 were found. All weld metal samples of the dissimilar joints met the Fig. 4 Microhardness mapping in cross-section: a S690QL- S355J2+N joint (t = 20  mm) welded with G69 and low heat input compared to b welded with high heat input, c S960QL-S355J2 + N joint (t = 20 mm) welded with G89 as single-V-groove weld compared to d double-V-groove weld, e S700MC-S355J2+N joint (t = 10 mm) welded with G79 as single-V-groove weld, and f as double-V-groove weld 2798 Welding in the World (2023) 67:2791–2802 1 3 requirements of their filler metal specifications. No relevant influence was found as a result of dilution with base metal of a significantly higher or lower strength class and differing composition [4]. The weld metal toughness properties of the single- V-groove welds t = 20 mm are represented by the impact energy values KV2 for VWT 0/1 specimens in Fig. 5b). The error indicators refer in each case to the minimum and max- imum values determined. The dark bars represent results from relevant test temperatures with regard to a procedure qualification. The pale bars show additional selected results from test temperatures in accordance with the filler mate- rial specification. All specimens met the minimum impact properties of 30 J at −40 °C specified for the base metals S690QL and S960QL [31]. As the strength of the weld metal increases, lower impact energy values were obtained. The highest impact energies of KV2 > 100 J were found for G46 weld metal, and lowest of about 40 J for G89 weld metal, respectively, confirming earlier investigations [12]. The force–deflection curves recorded in the instrumented test procedure predominantly indicated a failure in the transi- tion region with ductile content (Form E acc. to ISO 14556). Although the weld metal with higher strength reaches higher maximum forces, the ductile-to-brittle transition temperature of the HSS/UHSS are generally higher, so that increasingly lower impact energies were achieved at the same test tem- perature compared with the normal-strength steels. All CVN specimens for HAZ testing (VHT 1/1) also achieved the required minimum impact properties of the respective base material standard, i.e., for S355J2+N at least 27 J at −20 °C and for high strength QT steels S690QL/ S960QL at least 30 J at −40 °C [31], see Fig. 6. Due to the inhomogeneous material state with vary- ing proportions of HAZ and weld metal of the different filler metals in the ligament of the CVN test pieces, some- times strong scattering occurs within a series. A tendential dependence on the groove geometry was not found. For the S355J2+N side, sufficiently high impact energies are achieved even when using high-strength filler metals. For the S960QL side, it can be seen that higher impact energy values Fig. 5 Material properties of weld metal of t = 20 mm single-V-groove welds: a stress–strain curves, and b CVN impact energies for weld metal specimens of t = 20 mm butt welds compared with material requirements Fig. 6 CVN impact energies for HAZ specimens of t = 20 mm butt welds compared with material requirements 2799Welding in the World (2023) 67:2791–2802 1 3 result for welds made with G46 filler metal. This is due to the higher toughness level of the undermatching G46 weld metal itself, since there is always a proportion of it present in the ligament. Generally, half-V-groove joints would allow to investigate separate HAZ subregions more accurate, but they were not considered in this study. Only single-V and double-V-groove joints were investigated due to their better operative weldability and higher relevance in manufactur- ing practice. Based on the results of the destructive tests on the welded joints, a generally good weldability was found. In contrast to sophisticated rolling and heat treatment processes applied for base material manufacturing, the strength and toughness of the weld metal can only be achieved by solid solution and precipitation strengthening causing a higher concentration of refinement and hardening elements. Accordingly, metal- lurgical problems commonly result for these materials, i.e., sensitivity to excessive hardening, cold cracking, or solidifi- cation cracking [5, 9]. Nonetheless, neither the various mix- tures of the different material compositions nor the different welding parameters and weld preparations were identified as critical in the range of variations applied. These findings are limited to the performed experimental work, i.e., welding procedures, and scope of testing. Especially for other pro- cesses with significantly different weld metallurgy, such as submerged arc or laser beam welding, the weldability might behave in a fundamentally different way. 3.2 Load‑carrying capacity The maximum load-carrying capacity σmax of the welded joint was calculated as the maximum tension force in relation to the initial cross section S0 at the point of failure in the parallel length area of the test piece. The determined load-carrying capacity is compared with the ultimate tensile strength Rm of the weaker joint partner, which always was the S355J2+N. The results for the specimens of plate thick- ness t = 10 mm are shown in Fig. 7. The error indicators refer in each case to the minimum and maximum values determined. The maximum nominal stresses determined for the t = 10 mm specimens ranged from 558 to 576 MPa. Taking into account the measurement uncertainties and typi- cal scatter, this corresponds to the tensile strength of the S355J2+N. This matches with the typical failure behavior of the specimens. Under tensile loading, no relevant deformation occurred in the weld metal or soft zone of the HSS’s HAZ (see Fig. 8a). All specimens with t = 10 and 20 mm showed ductile failure in the unaltered base material (see rep- resentative fracture of a specimen in Fig. 8b). A local distribution of the deformation corresponding to the heterogeneous strengths in the weld joint region was not determined (cf. [17]). The determined load-carrying capacities of the speci- mens with plate thickness t = 20 mm are shown in Fig. 9. The determined load capacities are all in the same range of 526–537 MPa, and thus also correspond to the tensile strength of the normal-strength steel. The soft zone within the HAZ of the high-strength steels is thus not consid- ered relevant for the load-carrying capacity of the mixed joints with S355J2+N. It is not necessary to select the filler metal according to the higher strength steel connected. As expected, it is sufficient if the nominal strength (fy and fu) of the lower strength base metal is achieved. Fig. 7 Transverse load-carrying capacity of the dissimilar steel butt welds (t = 10  mm) compared with the ultimate tensile strength of the S355J2+N 2800 Welding in the World (2023) 67:2791–2802 1 3 4 Conclusions In this paper, multilayer MAG-welded mixed joints of the normal-strength steel S355J2+N in combination with an HSS/UHSS, viz., S690QL, S700MC, or S960QL, with a plate thicknesses of t = 10 and 20 mm, were investigated. The weldability was analyzed based on material property tests and the load-bearing behavior in the transverse tensile tests. Among other, the results provide the empirical basis for a new or extended design model for mixed joints up to S960. The following main conclusions are drawn: a) The weldability of the investigated welding procedures as a combination of base material pairing, filler metal, heat input, and geometric joint configuration is given. The metallurgical weldability is suitable, and the mate- rial properties in the weld area required for weld proce- dure qualification acc. to ISO 15614-1 [28] have been proved for the samples with a thickness of t = 20 mm. b) The HSS/UHSS underwent softening in the HAZ because of the temperature cycles during welding. The extent of this soft zone depended on the steel grade and the welding procedure, i.e., heat input and weld build- up. However, the average tensile strength in the soft zone of the high-strength steels S690QL, S700MC, and S960QL was always higher than the tensile strength of the unaltered normal-strength S355J2+N. c) For all dissimilar steel butt welds as connections of S355J2+N and HSS/UHSS, the tensile strength of the normal-strength steel was always achieved in the trans- verse tensile test. d) Failure occurred exclusively in the normal-strength steel. Also, no relevant strain was observed at the side with the higher strength joint partner. The soft zone formed in the Fig. 8 Typical failure behavior of the dissimilar steel weld joints under transverse tension: a local strain distribution at different test times with increasing load levels, and b exemplary broken specimens with double-V-groove weld, t = 10 and 20 mm Fig. 9 Transverse load-carrying capacity of the dissimilar steel butt welds (t = 20 mm) compared with the S355J2 + N ultimate tensile strength 2801Welding in the World (2023) 67:2791–2802 1 3 HAZ of the HSS/UHSS was identified as not relevant to the load-bearing behavior. Neither an influence on the load-carrying capacity nor on the failure behavior of the welded joint was found. No relevant deformation occurred in the soft zone. e) There is no need to select a filler metal with a strength oriented to the high-strength joining partner to ensure the full load-carrying capacity of the mixed joint. Acknowledgements The research project IGF 21412 BG/P 1507 'Effective design concepts for welded mixed connections in steel struc- tures' from the Research Association for steel Application (FOSTA), Sohnstraße 65, 40237 Düsseldorf, Germany, is a sub-project of the FOSTA joint research program HOCHFEST and was supported by the Federal Ministry of Economic Affairs and Climate Action the German Federation of Industrial Research Associations (AiF) as part of the programme for promoting industrial cooperative research (IGF) on the basis of a decision by the German Bundestag. Author contribution All authors contributed to the study conception and design. Sample manufacturing was contributed by O. B. and S. E. The experimental investigations and data analysis were carried out by O. B., M. v. A., and S. E. The first draft of the manuscript was written by O. B.. M. v. A. and S. E. commented on previous versions of the manuscript. K.-M. H., J. P. B., and U. K. helped supervise the project. J. P. B. and U. K. conceived the original idea. All authors discussed the results and commented on the manuscript. Funding Open Access funding enabled and organized by Projekt DEAL. Declarations Conflict of interest The authors declare no competing interests. Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. References 1. Scharff R, Siems M (2014) Monopiles der nächsten Generation – Aspekte der Nachweisführung. Stahlbau 88(S2):2–9. https:// doi. org/ 10. 1002/ stab. 20143 0002 2. Brätz O, von Arnim M, Eichler S, Gericke A, Henkel K-M, Hildebrand J, Bergmann JP, Kuhlmann U (2023) Load-carrying capacity of MAG butt and fillet welded joints on high-strength structural steels of grade S960QL and S960MC. Ce/Papers 6(3– 4):587–594. https:// doi. org/ 10. 1002/ cepa. 2478 3. Li D, Huang Z, Uy B, Thai H-T, Hou C (2019) Slenderness lim- its for fabricated S960 ultra-high-strength steel and composite columns. J Construct Steel Res 159:109–121. https:// doi. org/ 10. 1016/j. jcsr. 2019. 04. 025 4. Mvola B, Kah P, Martikainen J, Suoranta R (2016) Dissimilar high-strength steels: fusion welded joints, mismatches, and chal- lenges. Rev Adv Mater Sci 44(2):146–159 5. Tümer M, Schneider-Bröskamp C, Enzinger N (2022) Fusion welding of ultra-high strength structural steels — a review. J Manuf Proces 82:203–229. https:// doi. org/ 10. 1016/j. jmapro. 2022. 07. 049 6. Maurer W, Ernst W, Rauch R, Vallant R, Enzinger N (2015) Eval- uation of the factors influencing the strength of HSLA steel weld joint with softened HAZ. Weld World 59:809–822. https:// doi. org/ 10. 1007/ s40194- 015- 0262-z 7. Amraei M, Ahola A, Afkhami S, Björk T, Heidarpour A, Zhao X-L (2019) Effects of heat input on the mechanical properties of butt-welded high and ultra-high strength steels. Eng Struct 198:109460. https:// doi. org/ 10. 1016/j. engst ruct. 2019. 109460 8. Bayock FN, Kah P, Mvola B, Layus P (2019) Effect of heat input and undermatched filler wire on the microstructure and mechani- cal properties of dissimilar S700MC/S960QC high-strength steels. Metals 9(8):883. https:// doi. org/ 10. 3390/ met90 80883 9. Chen Z, Xiong Y, Qiu H, Lin G, Li Z (2018) Stress intensity factor-based prediction of solidification crack growth dur- ing welding of high strength steel. J Mater Process Technol 252:270–278. https:// doi. org/ 10. 1016/j. jmatp rotec. 2017. 09. 031 10. Zhang Y, Xiao J, Liu W, Zhao A (2021) Effect of welding peak temperature on microstructure and impact toughness of heat-affected zone of Q690 high strength bridge steel. Mater 14(11):2981. https:// doi. org/ 10. 3390/ ma141 12981 11. EN 1993-1-8:2005+AC:2009 (2009) Eurocode 3: Design of steel structures - Part 1-8: Design of joints. CEN, Brussels 12. Schneider C, Ernst W, Schnitzer R, Staufer H, Vallan R, Enz- inger N (2018) Welding of S960MC with undermatching filler material. Weld World 62:801–809. https:// doi. org/ 10. 1007/ s40194- 018- 0570-1 13. Cai W-Y, Wang Y-B, Li G-Q, Stroetmann R (2022) Comparative study on strength of TMCP and QT high-strength steel butt- welded joints. J Construct Steel Res 197:107447. https:// doi. org/ 10. 1016/j. jcsr. 2022. 107447 14. Zhang L, Kannengiesser T (2016) HAZ softening in Nb-, Ti- and Ti + V-bearing quenched and tempered steel welds. Weld World 60:177–184. https:// doi. org/ 10. 1007/ s40194- 016- 0299-7 15. Afkhami S, Javaheri V, Amraei M, Skriko T, Piili H, Zhao X-L, Björk T (2022) Thermomechanical simulation of the heat- affected zones in welded ultra-high strength steels: Microstruc- ture and mechanical properties. Mater Des 213:110336. https:// doi. org/ 10. 1016/j. matdes. 2021. 110336 16. Bergmann JP, Hildebrand J, Kuhlmann U, Spiegler J, Keitel S, Mückenheim U (2020) Tragfähigkeit von Stumpfnähten höher- fester Stähle im Stahlbau, Research report IGF 19470 B. DVS Media, Düsseldorf 17. Ran M-M, Sun F-F, Li G-Q, Kanvinde A, Wang Y-B, Xiao RY (2019) Experimental study on the behaviour of mismatched butt welded joints of high strength steel. J Construct Steel Res 153:196–208. https:// doi. org/ 10. 1016/j. jcsr. 2018. 10. 003 18. EN 1993-1-1:2005+AC:2009 (2009) Eurocode 3: Design of steel structures - Part 1-1: General rules and rules for build- ings. CEN, Brussels 19. DIN EN 1993-1-8/NA:2020-11 (2020) National Annex - Nation- ally determined parameters - Eurocode 3: Design of steel struc- tures - Part 1-8: Design of joints. DIN, Berlin 20. EN 1993-1-12:2007+AC:2009 (2009) Eurocode 3: Design of steel structures - Part 1-12: Additional rules for the extension of EN 1993 up to steel grades S700. CEN, Brussels 21. DIN EN 1993-1-12/NA:2011-08 (2011) National Annex - Nationally determined parameters - Eurocode 3: Design of steel http://creativecommons.org/licenses/by/4.0/ https://doi.org/10.1002/stab.201430002 https://doi.org/10.1002/stab.201430002 https://doi.org/10.1002/cepa.2478 https://doi.org/10.1016/j.jcsr.2019.04.025 https://doi.org/10.1016/j.jcsr.2019.04.025 https://doi.org/10.1016/j.jmapro.2022.07.049 https://doi.org/10.1016/j.jmapro.2022.07.049 https://doi.org/10.1007/s40194-015-0262-z https://doi.org/10.1007/s40194-015-0262-z https://doi.org/10.1016/j.engstruct.2019.109460 https://doi.org/10.3390/met9080883 https://doi.org/10.1016/j.jmatprotec.2017.09.031 https://doi.org/10.3390/ma14112981 https://doi.org/10.1007/s40194-018-0570-1 https://doi.org/10.1007/s40194-018-0570-1 https://doi.org/10.1016/j.jcsr.2022.107447 https://doi.org/10.1016/j.jcsr.2022.107447 https://doi.org/10.1007/s40194-016-0299-7 https://doi.org/10.1016/j.matdes.2021.110336 https://doi.org/10.1016/j.matdes.2021.110336 https://doi.org/10.1016/j.jcsr.2018.10.003 2802 Welding in the World (2023) 67:2791–2802 1 3 structures - Part 1-12: Additional rules for the extension of EN 1993 up to steel grades S700. DIN, Berlin 22. Spiegler J (2022) Tragfähigkeit von Kehlnaht- und Stumpfnaht- verbindungen höherfester Baustähle. Dissertation, University of Stuttgart, Institut für Konstruktion und Entwurf, Mitteilung No. 2022-2. https:// doi. org/ 10. 18419/ opus- 12207 23. FprEN 1993-1-8:2023 (2023) Eurocode 3: Design of steel structures - Part 1-8: Joints. Formal vote version. CEN/TC 250, Brussels 24. CEN/TC 250/SC 3/WG 12 N118 (2021) prEN 1993-1-12: Eurocode 3 - Design of steel structures - Part 1-12: Additional rules for steel grades up to S960. Working Draft (unpublished). CEN, Brussels 25. ISO/TR 17671-2 (2002) Welding - Recommendations for welding of metallic materials - Part 2: Arc welding of ferritic steels. ISO, Geneva 26. DIN 50159-1 (2022) Metallic materials - Hardness testing with the UCI method - Part 1: Test method. DIN, Berlin 27. von Arnim M, Eichler S, Brätz O, Hildebrand J, Gericke A, Kuhlmann U, Bergmann JP, Flügge W (2022) Effiziente Nach- weiskonzepte für geschweißte Mischverbindungen im Stahlbau. Stahlbau 91(10):660–670. https:// doi. org/ 10. 1002/ stab. 20220 0046 28. EN ISO 15614-1:2017+A1:2019 (2019) Specification and quali- fication of welding procedures for metallic materials - Welding procedure test - Part 1: Arc and gas welding of steels and arc welding of nickel and nickel alloys. CEN, Brussels 29. ISO 9016 (2022) Destructive tests on welds in metallic materi- als - Impact tests - Test specimen location, notch orientation and examination. ISO, Vernier 30. ISO/TR 581:2005 (2005) Weldability - Metallic materials - Gen- eral principles. ISO, Geneva 31. EN 10025-6:2019+A1:2022 (2022) Hot rolled products of struc- tural steels - Part 6: Technical delivery conditions for flat products of high yield strength structural steels in the quenched and tem- pered condition. CEN, Brussels Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. https://doi.org/10.18419/opus-12207 https://doi.org/10.1002/stab.202200046 Mechanical properties of MAG butt welded dissimilar structural steel joints with varying strength from grade S355 up to S960 Abstract 1 Introduction 2 Experimental methodology 2.1 Material and sample welding 2.2 Weld formation and hardness distribution 2.3 Mechanical weld testing 3 Results and discussion 3.1 Weldability 3.2 Load-carrying capacity 4 Conclusions Acknowledgements References