Applied research of high-strength steel utilization for a track of demining machine in terms of mechanical properties

The aim of this paper is to investigate welded joints of high-strength steel S960 QL manufactured by using three different welding technologies, namely the electron beam, the laser beam, and the metal active gas (MAG) technologies. The experimental part included tensile strength evaluation, microstructural analysis of welded joints, and hardness measurement. Welded joints (WJ) have consisted of the identical steels and the identical thickness (10 mm). Destructive tests confirmed that welded joints are characterized by the tensile strength similar to the base material. Upon further observation, we can conclude that microhardness was characterized by the lowest value in the softening zone (SZ) and the highest value in the hardening zone (HZ). The degree of softening was 11% for the electron welding, 13% for the laser welding, and up to 27% for the conventional MAG welding. This also corresponds with the size of the SZ, which was the widest in welds made with the MAG technology. The laser beam weld produced up to 50% lower heat-affected zone (HAZ) compared to the conventional MAG technology. In case of the electron beam, this number is even higher. On the contrary, highest hardness was observed for the electron beam technology, where the hardness in the hardening zone increased by up to 40% when compared with the base material. Tests show the possibility of production of reliable welded joints, which meet the complex requirements for lifetime and quality (according to the standard EN 6520–1 focusing on defects categorization and EN 5817 dealing with defects tolerance).


Introduction
All manufacturing materials and properties are subject to wear, aging, fatigue, or other adverse conditions. Since maintaining a healthy ecology is an important task in the twenty-first century, it is desirable to reduce an impact of material production (e.g., reduction of CO 2 production) and extend use beyond the intended service life. The production of the greenhouse gas is largely linked to both fuel consumption of various vehicles (and thus to their weight) and also to production of materials (most commonly steel) [1,2]. CO 2 is a by-product commonly occurring during steel production. Considerable efforts are underway to reduce the weight of structures and thus to make more efficient use of steel as an important material. Structural weight can be reduced by using high-strength steels. For the transport industry, high-strength steel is the fastest growing group of materials due to advantageous strength-to-weight ratio [3][4][5]. Recently, an implementation of advanced welding technologies in various industries is considered as one of the most important trends allowing modernization of technological processes [6,7].
The thermal cycles of welding technologies cause changes in the microstructure of steels, which lead to the formation of a melting zone (FZ) and a heat-affected zone (HAZ) in the joint of welded components. Because welding causes a high rate of heating as well as cooling, FZ is mainly formed by martensite [8][9][10][11][12][13][14][15]. The cooling rate in progressive welding methods (laser, electron) is usually higher than the critical cooling rate for the formation of a martensitic structure. For this reason, and thanks to alloying elements, almost all modern high-strength steels are characterized by high hardenability. Therefore, a martensitic structure is observed in their weld metals [14,16]. The formation of a melting zone has a significant influence on the hardness due to welding. The softening of the weld, which is observed far from the fusion line, has an adverse effect on the mechanical properties thereof. Significant research has been conducted to address this condition [17]. This paper is a continuation of the extensive research into the optimization of the crawler belt of a military demining machine "Božena 5" (Fig. 1). The main goal is to evaluate the hardness of base material as well as hardness of weld joints created by the metal active gas (MAG) technology, the laser beam, and the electron beam welding technology for high-strength steel S960 QL. It has been previously demonstrated that progressive welding technologies do not reduce the tensile strength or the yield strength of the base material [18]. On the contrary, the conventional MAG technology reduces the yield strength by about 20%. Another important finding is the fact that the average values of transverse residual stresses on the surface of the material after welding reach only 250 MPa in tension for laser welding, and for electron welding, the stress was reduced to 50 MPa in tension, and compressive stress was observed in electron welding [19]. All of the presented findings play an important role in the optimal design of the crawler belt of the remote-controlled demining machine in terms of durability and safe operation thereof. The environment, in which the machine operates, is extremely lifethreatening (minefields).
Hardness modification and evaluation broadens the horizons of the presented research. To further elucidate, reference [18] has shown that materials welded by progressive welding methods reveal significantly lower adhesive-abrasive wear in operation compared to the conventional MAG technology. At the same time, a decrease of the toughness of welded joints created by progressive welding methods was observed. Therefore, it is reasonable to continue research to investigate, how individual welding technologies will affect the hardness of the material, which plays an important role in the complex interpretation of the results.

Problem statement
Expediencies of the demining machine Božena 5 producer (Way Industries, Inc.) show, that application of scientific knowledge even for simply particular tasks causes difficulties. It seems that the elaboration of the scientific research set [20,21] would be useful as a bond of base research and applied research for the use of their results directly in the practice. Applied research can be effectively developed only if the base research is performed well in advance. This meets issues. Currently, the individual general problem of highstrength steels is well managed. However, possibilities of optimization are offered in the issue of lifetime of welded high-strength structures and their implementation into complex loaded structural units.
A proper path of every research is in an economical approach, which is based on sources. Therefore, every designed component has to be analyzed in detail. This also applies to the belt of the demining machine Božena 5. Evaluation of hardness performed in this contribution broadens horizons of the entire presented research. It has been proven [21] that the material, which is welded by progressive welding technologies (laser, electron), has significantly lower operational wear in comparison with the conventional welding technology MAG. Therefore, it is reasonable to continue in the research to identify how individual welding technologies influence material hardness. Odds and irregularities (e.g., send between a belt welded joint and a rosette) lead to a significant contact pressure increase and to a formation of plastic deformations. These findings have been published in [21,22].
For each material pair, the values of contact pressures in the contact area of irregularities increase many times over. Normal stress reached in an ideal friction surface does not cause the yield of strength to be exceeded (in case of the Božena 5, this is the contact of a rosette or pulleys with a belt) (Fig. 2).
If odds and irregularities (sand, stones, etc.) are taken into account, local areas with exceeded reduced stress may appear. Then, areas with plastic deformations begin to form. This phenomenon leads to one of the possible types of wear, which may origin in the contact. Investigation of adhesive-abrasive lifetime of the tested material S960QL and its welded joints (electron, laser, MAG) has been presented in [21]. Belt tearing of the demining machine occurs when operating the machine under specific conditions. Diagnosis, analysis, and subsequent resolution of this problem are extremely important not only for the manufacturer of the remotely controlled vehicle (Way Industries, Inc.), but also for a general contribution to scientific and research information and findings. Analysis of the current state of the vehicle is briefly presented here (details are provided within the allowed scope of military documentation).
The drive unit of the "Božena 5" (Fig. 3a) is supported by four hydraulic motors, which serve as a power unit for the front drive wheels (1) and also for the drive sprocket (rosette) (2). Driving is the standard as with the most skid steer loaders. Belts (4) can be braked separately, which enable a change of movement (rotation). It is also possible to turn the machine while almost completely stationary by moving one belt to one side and the other belt to the opposite direction. Each of these movements causes a variable strain on the belt. By repeating these movements cyclically, the belt is subjected to fatigue. Of course, only rosettes are sufficient for movement with a given crawler chassis. However, when moving on wheels (Fig. 3b), it would not be possible to change direction. Therefore, the front wheels must also be powered by motors, which allows a change to a wheel chassis at the same time. To be able to achieve this change, it is necessary to replace the rosette with the inserted (middle) wheel (3). An important part of the chassis which is necessary for belt tensioning is performed by using tensioning pulleys (5) operated by a screw mechanism.
The wheel hub shaft is made of 34CrNoMo6 alloy. The wheel disk and other chassis components are made of S355J2 alloy, which represents structural steel with good weldability and yield strength R e = 320-360 MPa. After experimental optimization, the chassis belt will be manufactured using S960 QL steel, characterized by a high hot forming strength. The yield strength of this material is 960 MPa. The main reason for using high-strength steel is to lighten the unstrung parts of the chassis, causing the belt to be more lightweight while maintaining necessary strength. Authors are constantly conducting research on the suitability of this material (fatigue wear, tribological wear, welding technology, residual stresses, microstructure, hardness, FEM analysis, etc.) [18,19].
To give an idea of the total belt load, it can be mentioned that the maximum total weight of the machine is 13,090 kg. The weight of the machine itself is 10,240 kg and the weight of the attached earth milling cutter is 2850 kg. However, in addition to the self-weight load and the resulting driving resistance, all possible load methods and combinations must be identified.
The research presented in this paper also consists of measuring the hardness of the material affected by welding technologies. At this moment, it is possible to theoretically and generally analyze the effect of hardness on previously obtained results of fatigue measurement [20] and tribological lifetime [21]. In the work of Sága et al. [20], the authors conducted a study, which analyzed the fatigue life of the base material S960 QL and its welds using the technologies in question. From there, it can be argued that progressive welding technologies (laser, electron) had only a minimal effect on the fatigue life in comparison with the base material. The current work expands the previous research on fatigue measurement by performing a series of static tensile tests (among other things) in order to confirm that even this mechanical property will not be affected by progressive technologies (yield strength and tensile strength will not decrease). If it were possible to faithfully describe the relationships between fatigue life, stress state differences, and damage, then the strength of the welded material could serve as a measure of fatigue life prediction.
Even without carrying out tensile tests, it is possible to state the postulate of maintaining equivalence between the residual strength after welding and the static strength determined by the tensile test. It goes without saying that the residual strength will gradually degrade in the process of fatigue testing. Degradation can only continue until the state of reaching the maximum possible load, which will cause failure. Thus, if welding causes a decrease in strength compared to the base material, this is a way to partially explain its adverse effect on fatigue life. Partly because there are far more input factors and marginal conditions that affect the life of the material. For example, in the work of Blatnický et al. [21], the authors conducted a study in which they determined the amount of adhesive-abrasive wear of the base material S960 QL and its welds by the appropriate welding technologies. During the testing, the mechanism of material strengthening and the wear decreased in real time. It was found that the MAG welds achieved the highest wear of all the samples tested. The explanation was the reduced value of the strength and ductility of the MAG weld in conjunction with the increasing thermal performance of the thermomechanical effect in the HAZ. Thus, a lower welding speed with high heat input will affect a larger area. This is subsequently manifested in a more pronounced deformation and expansion effect when cooling the welded material. This is the main reason for the faster wear of the material in the adhesive-abrasive way of wear. Different hardness ratios between the tested samples (base material, welding technologies) and the frictional counterpart will cause a different distribution of abrasive and adhesive wear. A low difference in the hardness of contacting bodies (and hard) will cause a greater impact of the abrasion mechanism. Conversely, with a large difference, the adhesion mechanism will have a greater impact on wear.
The results of welding hardness measurements can be used in a future study to determine the optimal ratio of the adhesive-abrasive wear method. This ratio can be used to achieve the desired state of service life of the demining machine belt stressed by the adhesive-abrasive method of wear. Therefore, the results of the measurements resulting from the presented paper will be important inputs in the optimization of the final product.

Material
The steel used in this research is Strenx 960 steel without additional designation. It was delivered in the form of 10-mm thick plates. It is steel for which the manufacturer (SSAB) guarantees compliance with the requirements of the EN 10025-6/S960QL standard (Table 1). Steel is a highstrength hot-rolled structure material (hardened and tempered) (Q) while maintaining its properties at low temperatures (L).
The manufacturer (SSAB company) states the chemical composition prescribed in the delivery note according to Table 2.
The actual chemical composition of an evaluated series (performed on a spectrometer in an external company) is given in Table 3. Table 4 shows the measured mechanical properties determined by the tensile test and the impact test according to STN EN ISO 6892-1. Table 1 The chemical composition of the tested material is determined by the EN 10 025-6 standard [23] Steel Chemical composition-max wt.

Welding method and equipment
Carbon equivalent can be determined in two ways, either according to method AC.2-carbon equivalent CE(CEV) (1) or according to method BC.3-carbon equivalent CET (2). Method AC.2 is used for steels of classical production and method BC.3 is used for microalloyed fine-grained highstrength steels. The main emphasis is placed on the elimination of hydrogen (cold) cracking.
Carbon equivalent was calculated on the basis of the given equations: CEV = 0.54 wt. % and CET = 0.359 wt. %. These values theoretically indicate the need for preheating before welding. However, the steel manufacturer recommends preheating (75 °C) for this material up to thicknesses greater than 10 mm. Therefore, preheating was not applied before welding. Welding (MAG, electron beam, laser beam) of steel plates was carried out in an external company (First Welding Company) with the demand to achieve the required condition of optimal welds in terms of strength. The MAG welding technology was performed automatically by the OTC DAIHEN welding robot (Fig. 5). Welding parameters for the MAG are summarized in Table 5.
Based on the above data (also data in Tables 6 and 7) and Eq. (3), the heat input during welding was calculated for individual methods.
where k is the thermal efficiency coefficient of the welding process, for MAG it reaches the value k MAG = 0.8 ( −) and for electron and laser welding k beam = 0.95 ( −).
The laser welding technology was performed using a YLS 5000-S1 fiber solid-state laser (Fig. 6). Figure 6b (1)  Table 4 Evaluated mechanical properties of S960 QL steel shows the welding process of the production of the test plates. A welding machine moves in guidance (towards a reader). Plates are positioned without a gap and mounted by means of holders to avoid deformation. The used welding parameters of this technology are summarized in Table 6. The electron beam welding process was performed with the PZ EZ JS30 JUMBO welding complex (Fig. 7). The used welding parameters of this technology are summarized in Table 7.
The hardness of a homogeneous material can be considered approximately constant. However, heat introduced into the material during welding causes changes of material properties [25]. This can greatly affect the behavior of the material [26]. Hardness measurement used causes changes of material properties [25]. This can greatly affect the behavior of the material [26]. Hardness measurement can be used for initial input, based on which many material properties can be determined. These include not only the yield strength, tensile strength, and fatigue, but also the distribution of residual stresses or fracture toughness. In practice, Eq. (4) is most often used for the relationship between yield strength and hardness:     where F is the test load and d 1,2 is the arithmetic mean of the two diagonals of the indentation. The measurement was performed on a Zwick/Roell ZHVµ automatic measuring device (ZwickRoell, Ulm, Germany).

Tests of welded joints
The research has included these tests: -Evaluation of tensile strength of welded joints for individual welding technologies (MAG, laser beam, electron beam) and it was compared with the base material (S960QL). The tensile test was performed on a universal hydraulic test device INOVA (Fig. 8a) according to the ISO 6892 standard. Test sample dimensions and design are depicted in Fig. 8b. -There was an investigation on the microstructure of the base material ( Fig. 9)  total width of the measured area 10 mm for a narrow HAZ (electron and laser technologies) was chosen, i.e., 5 mm for each side of the welded joint. The distance between individual imprints (Fig. 9a, b) was chosen of 250 µm to ensure identification of changes in all areas and transition zones. The HAZ is significantly wider for the MAG technology. Therefore, a measured area of 10 mm for one side from a welded joint was chosen with consideration of relatively symmetrical hardness results (Fig. 9c).
Hardness measurement has been performed in the lateral direction of welded joints (Fig. 9a-c). A matrix of coordinates of all points has been created, in which the hardness has been measured automatically. Totally, 16 hardness measurements for each welding technology have been performed. Samples have been produced from middle parts of weldments (Fig. 4).

Weld formation
Visual inspection disclosed that all produced welding joints have had smooth welded joint area without spraying, with a proper weld root shape (Fig. 10). In all three investigated welding processes (laser, electron, MAG), the microstructure of welds consists of three main parts. These are the base material, the heat-affected zone, and the weld metal. As it can be seen in Fig. 11a, the electron beam weld and consequently heat-affected zone of the material have a smaller width than the laser beam weld in Fig. 11b. However, the laser beam technology also guarantees a significantly narrower heat-affected zone and the Fig. 9 A A matrix of coordinates of hardness measurement of an electron beam in a cut section, b a matrix of coordinates of hardness measurement of a laser beam in a cut section, and c a matrix of coordinates of hardness measurement of a MAG in a cut section Fig. 10 Welded joint (having a thickness of 10 mm) viewed from the face side: a electron, c laser, and e MAG and the root side: b electron, d laser, and f MAG after the welding process weld itself when compared to the MAG welding (Fig. 11c), wherein the welded joint was made in two steps. The entire welded joint as well as the HAZ is considerably wider in comparison with the evaluated progressive technologies. Comparison of microstructure has uncovered no deficiencies of used welding technologies neither in HAZ nor in the welded joint area.
Metallographic samples after etching revealed clearly visible boundaries between the weld metal, HAZ, and the base material. Thanks to Fig. 11, it is possible to understand the geometric parameters of welds (width of the weld surface, height of the weld, width of the weld root).
The analysis of the weld destructive tests consisted of a visual inspection for fatigue (after breaking the sample). The whole process of fatigue testing was explained in detail in the paper [21]. The samples were broken in the weld metal (instead of the stress concentrator), so it was possible to perform a macrostructural examination of the weld crosssection and determine the presence of welding defects. Figure 12 shows welded samples for electron (Fig. 12a, c) and laser (Fig. 12d) after uniaxial torsion fatigue tests (Fig. 12a, b) and after uniaxial bend fatigue tests (Fig. 12c, d).
The fracture surfaces did not show the presence of defects caused by welding. The same conclusion was reached in the destructive tests using the tensile strength evaluation. Fractures of the tested samples occurred outside the welded joint (Fig. 13). Therefore, the presence of defects in the weld is not expected. These would cause a significant reduction in tensile strength, which has not been proven during the tests.

Microstructure characteristics
Microstructure of the base material (Fig. 14) consists of finegrained tempered martensite and tempered bainite. Grain refinement and strength increase due to precipitation hardening, which are achieved by precipitates of types Nb and Ti (C, N).
The microstructure after welding by all used methods (Figs. 15, 16, 17, 18, and 19) consists of three main parts: the base material, the HAZ, and the weld metal. The microstructure of the MAG weld metal (Fig. 15) consists of martensite. The martensitic microstructure is achieved mainly by the chemical composition of the additive material rich in Mn, Ni, and Cr.
The laser beam-welded metal microstructure (Fig. 16) includes coarse-grained martensite formed by remelting the base material. Grain size is mainly affected by high temperatures and holding time, which caused a coarsening of austenite grains. Figure 17 shows the fine-grained structure of the weld metal and the transition from the weld metal through the HAZ to the laser-welded Strenx 960 base material. It can be seen that the HAZ has three areas. In the subcritical HAZ, the temperature during welding was not higher than Ac 1 (eutectoid temperature during heating). Martensite in the base material is tempered. During the evaluation of hardness, softening of the HAZ can be expected. In the intercritical HAZ, the temperature during welding ranged between Ac 1 and Ac 3 (conversion temperature of Fe α -ferrite to Fe γ -austenite during heating). The basic material begins to form austenite. In the supercritical HAZ, the base material is austenite, and depending on the degree of superheating of the material above the Ac 3 temperature, undesired growth of austenite grains may occur. After cooling, martensite is usually formed again (depending on the cooling rate and the quantity of alloying elements).
The HAZ in Fig. 17 from left to right is as follows: the base material (BM); the SCHAZ (sub-critical HAZ) zone formed by the tempered base material; the ICHAZ (inter-critical HAZ) zone formed by a mixture of tempered BM and a mixture of decay structures (M, B, F-due to low temperature, no complete austenitization occurred; not all carbon dissolved in austenite) with temperature range in the given area between Ac 1 and Ac 3 ; the FGHAZ (fine-grained HAZ) zone with the fine-grained structure-temperature just above Ac 3 ; the CGHAZ (coarse-grain HAZ) zone with a coarse-grained structure. The structure is the coarsest in places with the highest Fig. 12 Fracture surfaces of test samples for investigation of fatigue lifetime evaluation: a the electron-welded joint loaded by the uniaxial torsion, b the laser-welded joint loaded by the uniaxial torsion, c the electron-welded joint loaded by the uniaxial bend, and d the laser-welded joint loaded by the uniaxial bend  temperature reached, just at the construction boundary and gradually decreasing with maximum temperature; the grain size decreases (all due to the increase of austenite grains)temperatures high above Ac 3 ; and the last (right side) is a weld metal with a coarse grain due to the high temperature. All of this is also applied to Fig. 13, but it is described from the opposite side. Figure 18 shows the structure of a Strenx 960 electron beam welded material. It is possible to observe the weld metal cooling direction and the formation of colloidal crystals in the direction of the reverse temperature gradient. Weld metal is coarse-grained due to the high temperature composed of martensite. Figure 19 shows the transition from the weld metal through the HAZ to the base material (from left to right).
The HAZ consists of two parts; the zone closer to the weld was heated above the Ac 3 temperature, but the temperature was not high enough to cause a greater growth of austenite grains and a fine-grained structure can be seen.

Mechanical properties
Tensile strength results are shown in Fig. 20; the graph represents the average value of the results from two measurements for each sample. Achieved results point to the fact that laser and electron beam welds have minimal effect on the strength of the material.
Samples were broken in all cases for laser and electron beam welded samples in the base material (only for MAG in  (Table 4). However, in all three cases (base material, laser, and electron beam weld metal), the material has a lower ductility than stated by the manufacturer. The average value of ductility from the six measurements is 11.32%, while the manufacturer states 15%.
On the contrary, the arc weld affected the strength of the material considerably. The samples were broken at the weld metal. The yield strength did not even reach the minimum yield strength of the base material. The yield strength decreased by about 20%. There was also a decrease in strength by about 15%. The value for yield strength to ultimate tensile strength ratio has changed from 0.96 ( −) to 0.88 ( −). In addition, the material also showed very low ductility, on average 4.3%. The determined average values of strength are clearly shown in Table 8.

Hardness distribution
For the electron and laser beam technologies, the total width of the measured area was 10 mm, since both technologies have a narrow HAZ. An average value of the observed hardness HV 0.5 for the base material is arithmetically 342 HV, which is equivalent to 331 HB (35 HRC) according to conversion tables. The total measured hardness maps of the base material S960 QL for one selected weld of each technology can be seen in Figs. 21, 22, and 23 and show all measured hardness values from 16 performed measurements for each technology.
The global maximum measured hardness values reach 470 HV and were observed in electron beam welds (Fig. 21a). They were located at the melted material boundary in the transition area, where the partially fused metal grains are located between the base material and the weld (a supercritical HAZ).
Examination of the laser-weld hardness map (Fig. 22b) revealed that the laser weld achieved relatively uneven hardness values directly in the weld metal. This is due to the overlap of the weld, which is in the middle of the welded plate. Hardness values range from 360 to 430 HV. In the    The HAZ width is much larger with MAG technology (Fig. 23). Therefore, and also due to the assumption of relatively symmetrical hardness results, a measurement area of 10 mm on one side of the weld was chosen. The average hardness in the MAG weld metal was 280 HV. Higher values of hardness can once again be observed at the boundary of the arc weld, with values ranging from 410 to 420 HV. However, these hardness values only reach a certain height above the weld root. This may be due to the fact that the weld was made in multilayer fashion. The HAZ of the base layer (at the root of the weld) was influenced by the subsequently applied cover layer of the second weld. After examining the overall hardness distribution map in Fig. 15, we found localized uneven decreases in hardness of the weld metal.

Discussion
Comparing the hardness results (Figs. 21, 22, and 23) of all three analyzed welding technologies (laser, electron, MAG), it was observed that the highest recorded hardness was achieved by using the electron beam welding (470 HV). The maximum hardness was located at the weld boundary, where partially molten metal grains are located. Identical conclusions were reached in [4,28,29]. The increased hardness in this area is related to the high proportion of martensite and the toughness of the heataffected zone [19,28]. When welding high-strength steels with high-performance electron beam with rapid heating and cooling, martensite is most often formed in the weld metal and the narrow HAZ is also advantageous. However, high strength martensitic steels are more prone to HAZ softening due to the original martensite content in the base material.
The electron weld (Fig. 21) demonstrated uniform hardness distribution in the weld metal. The hardness oscillates around the value of 435 HV. The lowest measured hardness value was of 307 HV. The microhardness trend for the electron weld was consistent and characterized by the highest value in the hardening zone and the lowest value in the softening zone as per [4,5]. The degree of softening in the critical zone in comparison with the base material was 11%. Softening appeared on the line near the weld surface, in the subcritical HAZ. The softening of the HAZ was directly related to the heat input during welding, which was explained in [30].
When decreasing heat input (MAG welding 1.67 kJ•mm −1 , laser welding 0.63 kJ•mm −1 , electron beam welding 0.55 kJ•mm −1 ), martensite had less time to temper in the base material, and in case of the electron beam welding, hardness has not been substantially reduced. The performed experiments have shown that the laser beam method is characterized by a narrower HAZ (Fig. 24b, c) in comparison with conventional method of arc welding. The width of the HAZ obtained by laser welding is reduced by up to 50% with given welding parameters. The overall width of the weld is reduced by up to 66%.
The hardness distribution for the investigated welding technologies can be seen in three selected lines (indicated Studies [4,5,15,17] were performed in order to assess the possibility of influencing the width of the weld when employing progressive technologies. Welding of highstrength steels using the MAG method with higher heat input causes the softening in the HAZ, which further affects the resulting properties of the material. The average hardness of the MAG weld was of 280 HV. This corresponds to 266 HB and tensile strength of 931 MPa in the static tensile test (see [19]). Further research [5,18,19,[31][32][33][34] has shown that weld geometry and microstructure cause crack initiation and propagation under cyclic loading.
Based on the obtained results, the design of optimal methodology for the construction of the "Božena 5" belt will be defined by using progressive welding methods. Authors in [19] have shown that the fatigue life of the material welded by progressive welding methods is minimally influenced, meaning that the material and a production technology for the machine belt is optimally determined from the point of view of fatigue stress. Problems can occur with the belt overload in case of simple tension (one of the analyzed belt loads in operation). Authors upon evaluating welded joints' tensile strength made by laser beam published in [35][36][37] have determined that the presence thereof does not cause any significant effects. This phenomenon can be explained by a narrow softening zone and a low degree of softeningthereby supporting the results of our experiment-where the width of the softening is significantly reduced. Therefore, it will be recommended to perform the welding at a higher welding speed and a lower beam power for achieving the optimal conditions. Based on the hardness results, it will be possible to determine the width of the weld softening and, based on research published in [17] as well as our findings, to influence the effect of the softening zone on the strength of the material after welding.
We presume that the degree of HAZ softening in all types of welding technologies was influenced by the content of the original martensite in the base material (small average deviations in the hardness of the softened areas of all welds). Measured differences in the minimum hardness values (307 HV for the electron beam, 299 HV for the laser beam, 252 HV for the MAG technology) are due to various temperature cycles as well as toughness values for individual welding technologies. According to [31,38], we can conclude that during the operation of the "Božena 5" machine, toughness will have a significant effect towards avoiding brittle fractures or cold cracks which occur when the belt is loaded (Fig. 25). The MAG technology currently employed in the production of belts causes a significant decrease in toughness compared to the base material and high-progressive welding technologies [5,19,39,40].
An example of damage in currently employed processes (using MAG welding) can be seen in Fig. 25. The damaged area is located near the weld surface close to the transition from weld metal to the supercritical HAZ. At the same time, the highest hardness values in the MAG weld were observed at this location. This structure will act as a tension concentrator during belt loading and can consequently lead to critical damage thereof. Based on the results achieved in the experimental work and at the same time in correlation with [20,21], it can be concluded that the laser beam technology is the optimal alternative for the given application.

Conclusions
Tests of welds (electron, laser, MAG) included the production and testing of butt joints. The joints (with a constant thickness of 10 mm) were made of S960QL steel. The tests revealed the possibility of obtaining suitable welded joints resistant to the demanding conditions resulting from the implementation in a mobile working machine. All produced welds were characterized by a smooth, uniform, and spatterfree weld surface with a weld root of the correct shape. Visual inspection and inspection of weld fractures did not Metallographic evaluation revealed that the laser structure of the laser-welded metal contains coarse-grained martensite formed by remelting the base material. The microstructure of the MAG weld metal contained a homogeneous structure of acicular martensite. The weld metal of the electron beam is coarse-grained; due to the high temperature, it is composed of martensite. No imperfections or defects were observed in the welded joints. The maximum hardness of the S960 QL material was reached during electron beam welding when the lowest heat input was used. Hardness in the hardening zone increased with the electron beam technology by up to 40% in comparison with the base material. The welded joint made by the laser beam demonstrated 50% lower HAZ in comparison with the conventional MAG technology. This number is even higher in case of the electron beam technology. By choosing the right welding technology, it is possible to suitably affect the tribological service life of welds by adjusting the degree of hardness without the need of additional heat treatment. The achieved results confirm that the electron beam technology is the optimal welding technology for the above-mentioned research. The resulting narrow softening zone and low degree of softening will have a minimal impact on the service life in the event of belt overload in service, tribological life, residual stresses, and also fatigue life.

Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, 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:// creat iveco mmons. org/ licen ses/ by/4. 0/.