EP2825345A1

EP2825345A1 - Method of producing a continuous metal strip by laser welding using a filler wire

Info

Publication number: EP2825345A1
Application number: EP13712517.5A
Authority: EP
Inventors: Steven Celotto; Tonny Van Der Veldt; Engbert Jan De Wit; Nick Johannes DEN UIJL
Original assignee: Tata Steel Nederland Technology BV
Current assignee: Tata Steel Nederland Technology BV
Priority date: 2012-03-14
Filing date: 2013-03-14
Publication date: 2015-01-21
Also published as: WO2013135847A1

Abstract

The present application relates to a method of producing a steel strip by laser welding the trailing edge (2) of a first steel strip (1) to the leading edge (4) of a second steel strip (3) wherein at least one of the steel strips to be welded together is an electrical steel strip, and wherein a filler wire is used

Description

METHOD OF PRODUCING A CONTINUOUS METAL STRIP BY LASER WELDING USING A FILLER WIRE

This invention relates to a method of producing a steel strip by laser welding the trailing edge of a first steel strip to the leading edge of a second steel strip.

Many processes in the metal industry, particularly the steel industry, originated as batch-type processes wherein a hot-rolled coil was treated as a coil during the subsequent processes of pickling, cold rolling, annealing and temper rolling.

Over the years, many of these processes have evolved into continuous processes wherein coils are joined together to form a continuous metal strip which is fed into the process. Continuous annealing, continuous cold rolling and continuous coating are examples of this evolution. Continuous pickling has been known for decades.

These continuous process require that metal strips, which are usually provided in the form of coils, that are joined together to form a continuous metal strip. A metal strip is fed into the process and the trailing end of this strip is joined to the leading end of a subsequent strip. This process of joining strips or coils can be repeated endlessly thereby forming a continuous metal strip for further continuous processing. While joining, the location where the joint is made is usually kept stationary while the joint is being made, while an accumulator is usually provided to assure undisturbed travel of the material to be further processed downstream of the joining device.

Several methods have been proposed for joining the trailing end of a first metal strip to the leading end of the subsequent metal strip. Mechanical methods such as dovetailing or seaming are generally considered unreliable or time- consuming and as a result, butt-joining by welding the coils together has become the most widely used.

The use of resistance seam welding and flash butt-welding has been proposed. However, these welding techniques have the problem of a local increase in thickness at the location of the joint because of the nature of the welding techniques that in the case of flash-but welding, requires deburring. EP-0 659 518- Bl proposes butt-joining by welding the metal strips using a laser beam to avoid the local increase in thickness at the location of the joint.

The disadvantage of using a laser beam like in EP-0 659 518-B1 is that the preparation of the trailing end of the first metal strip and the leading end of the subsequent metal strip has to be done very carefully. The quality of the shearing has to be very good because the quality of a laser weld is very sensitive to misalignment of the two parts to be joined. If the alignment is not adequate, the weld may be of insufficient quality, and it may break during continuous further processing, leading to an entire line stop until the strip is re-threaded and pulled through as well as to rejected material. This has a severe impact on the productivity of such a line and it may moreover lead to damage to the line or components thereof.

Maintenance of the shearing device is critical, because inadequate shearing may lead to inadequate welds. Regular maintenance of the shearing device adds to the maintenance costs of the line both in terms of money spent and standstill time.

It is an object of the invention to provide a method for butt-joining and welding metal strips together to produce a continuous metal strip using a laser beam, which produces a weld of consistently good quality which is less dependent on the quality of the alignment of the metal strip ends to be joined.

It is also an object of the invention to provide a method for butt-joining and welding metal strips together to produce a continuous metal strip using a laser beam, which provides a better first-time-right performance.

It is also an object of the invention to provide a method for butt-joining and welding metal strips together to produce a continuous metal strip using a laser beam, which is less sensitive to a poor strip entry shape.

It is also an object of the invention to provide a method for butt-joining and welding metal strip together to produce a continuous metal strip using a laser beam, which enables a reduction of maintenance costs and standstill costs due to maintenance or repair of the shearing means providing the cut metal strip ends to be welded together.

It is also an object of the invention to provide a method for butt-joining and welding metal strips together to produce a continuous metal strip using a laser beam from metal strips which are difficult to weld.

At least one of these objects is reached by providing a methodof producing a steel strip by laser welding the trailing edge of a first steel strip to the leading edge of a second steel strip wherein a filler wire is used and wherein the laser welding is performed by a laser that provides

(i) . a single spot power distribution, or

(ii) . a multimodal welding power distribution, wherein the multimodal welding power distribution has at least a first power peak (A) and a second power peak (B) separated from the first power peak by a power trough (C), wherein the first power peak converges on the trailing end of the first steel strip and wherein the second power peak converges on the leading end of the second steel strip, wherein at least one of the steel strips to be welded together is an electrical steel strip, wherein the pool of molten material created during welding has a composition comprising a mixture of the molten filler wire and the first and second steel strips and wherein the composition of the pool of molten material is such that after solidification of the pool of molten material, the solidified material becomes a fully austenitic microstructure at a point in time and that the said fully austenitic microstructure subsequently is transformed into a martensitic microstructure.

In the method according to the invention, in a method of producing a steel strip by laser welding the trailing edge of a first steel strip to the leading edge of a second steel strip, a filler wire is used. The filler wire is a metal wire which is added in the making of a joint. The filler wire is fed into the neighbourhood of the laser beam where it melts by the heat of the laser beam and becomes part of the pool of molten material which after solidification forms the weld. In the invention at least one of the steel strips to be welded together is an electrical steel strip. Electrical steel strips are difficult to weld together and simultaneously provide good bending properties. These properties are required to be of a minimum value to enable the welded strip to be passed through the installations, downstream of the welding process.

By using a filler wire, the process window for coil to coil joining is widened. Even though the trailing end of a first coil and the leading end of a second coil are cut and prepared for butt-welding, it is clear that in operational circumstances the cut edges may not always be perfect, or the strip shape may not be perfect, and as a consequence the match between the two abutting edges may not be perfect either. Poor abutment or alignment may be the result of different or poor strip entry shape, such as waviness, of the two metal strip ends to be joined. In the context of this invention the leading edge and the trailing edge are defined as the abutting edges. There may be small gaps between the abutting edges and when using a laser beam without the use of a filler wire the resulting weld will not be of a consistent quality over the width of the abutting edges and may even show holes where the laser beam was not able to form a melt pool. By adding the filler wire to the welding process, the imperfections in the abutting edges are compensated and a good and closed weld is obtained all along the length of the abutting edges.

According to the invention it is essential that the pool of molten material created during welding has a composition such that during cooling after solidification, the solidified material becomes a fully austenitic microstructure at a point in time during the cooling of the solidified material to ambient temperatures and that during that cooling the said fully austenitic microstructure subsequently transforms into a martensitic microstructure. The material of the pool of molten material (i.e. the composition of the melt pool) is a mixture of the molten material of the filler wire and the molten material of the first and second steel strips. The presence of the martensite ensures that the weld is sufficiently strong and tough to allow multiple bending operations before failure of the weld, and the presence of the 100% austenite during solidification is needed to break up the original large grain columnar ferritic microstructure that formed upon solidification. The composition of the melt pool is such that the natural cooling after the weld is sufficiently fast to allow the austenite to transform to a martensitic microstructure. With a martensitic microstructure in the context of this invention a microstructure is meant which contains at least 80% of martensite. Preferably the microstructure contains at least 90% martensite, preferably at least 95% martensite, more preferably only martensite. The martensite should not be brittle because this leads to rapid failure upon bending. It is therefore preferable that the carbon content of the martensite phase is kept below 0.4 wt.%. This is to be achieved by using low carbon filler wire. The carbon content of electrical steels is generally below 0.1%. If the natural cooling rate of the weld pool is insufficient to transform the austenite into the required amount of martensite, then the cooling of the weld can be sped up by accelerated cooling using e.g. air cooling, mist cooling or water cooling.

The laser welding is performed by a laser that provides a single spot power distribution or by a laser that provides a multimodal welding power distribution, wherein the multimodal welding power distribution has at least a first power peak (A) and a second power peak (B) separated from the first power peak by a power trough (C), wherein the first power peak converges on the trailing end of the first steel strip and wherein the second power peak converges on the leading end of the second steel strip.

Whether during solidification of the pool of molten material a fully (= 100%) austenitic microstructure is obtained at a point in time can be determined by using JMatPro^® (a registered trademark of Sente Software Ltd) and perform equilibrium calculations of the phase distribution as a function of time. If the austenite components shows a value of 100% in the time-phase diagram, then this requirement is deemed to have been met within the context of this invention. JMatPro^® is a simulation software which calculates a wide range of materials properties for alloys and is particularly aimed at multi-components alloys used in industrial practice.

The method according to the invention is particularly suitable for welding steel strips or coils together to form a continuous metal strip for processing in a continuous processing line such as a continuous pickling line, continuous rolling line, continuous annealing line, continuous coating line, continuous painting line and the like.

The laser energy may be provided by one or more single beam laser sources, or a multiple beam laser source. The multimodal welding power distribution may be provided by splitting the single or multiple laser beam using suitable optics. A single laser beam may for instance be split into two or more laser beams using splitter and mirrors. The weld between the first metal strip and the second metal strip is intended to join the abutting edges together to provide a continuous metal strip. With the method according to the invention a wider weld is provided due to the presence of a plurality of power peaks. The convergence or focussing of at least a first power peak on the trailing edge of the first metal strip and at least a second power peak on the leading edge of the second metal strip provides a power input over a wider distance than a single spot weld. Moreover, when using a single spot weld, i.e. a welding power distribution comprising only one focussed laser beam, the power is concentrated on the location where the trailing edge abuts the leading edge. When using the multimodal weld power distribution, the power is introduced in the abutting edges to be joined immediately next to the location where the edges abut or the gap if the edges do not abut perfectly, thereby achieving a more consistent power transfer. The addition of a filler wire further improves the quality of the weld and widens the operations window even further. Care has to be taken to introduce the filler wire at the right location and the right time to avoid that the filler wire does not melt properly. If the filler wire is introduced in the power trough at a moment when the power in that trough is insufficient, then a poorer weld quality will be the result. Therefore a proper tuning of the power peaks, both in power and in location, with the timing of the addition of the filler wire as well as the thickness of the filler wire are important process parameters that need to be controlled carefully. The advantage of a multimodal laser beam is that the weld quality is not affected by the presence of a gap.

It should be noted that the location of the power trough can be chosen by a suitable adjustment of the optical arrangement for focussing and/or splitting a laser beam or by providing one or more additional beams. This way, the depth of the trough can be adjusted between very deep (i.e. substantially no welding power in the trough) and almost absent (i.e. the welding power in the trough is at most the welding power of the first or second peak).

The movement of the laser-beam along the abutting edges is to be understood in such a way that the laser beam moves substantially along or parallel to the abutting edges so as to appropriately provide the welding power to the metal strip edges to be welded together to produce a continuous metal strip. This movement is a relative movement of the metal strips with regard to the laser beam. Moving the metal strip edges to be welded together along a stationary laser beam is therefore also embodied in this invention, as well as the simultaneous movement of the metal strip ends to be welded together along a moving laser beam.

If the metal strip ends to be welded together are not cleanly sheared due to wear of the shearing device leading to poor abutment or misalignment (horizontal and/or vertical), the multimodal weld power distribution in combination with the proper use of a filler wire ensures that a proper weld of good quality can be made, despite any misalignment that may be present. This relative insensitivity for misalignment can be translated in a reduced need for maintenance or repair of the shearing device, leading to a longer life span of a set of knives, and thereby in lower maintenance costs and standstill costs. Typically, after prolonged use, the quality of the shearing diminished due to wear of the shearing device and the knives. As a result, the edges of the metal strips where the joint is to be made, start to show vertical and/or horizontal misalignments resulting in a significant decrease in weld quality, resulting in a low first time right efficiency when using a single spot laser welder. Since an apparatus providing a multimodal welding power distribution can deal much more easily with horizontal and vertical misalignments, the life span of the shear cassette can be increased.

In an embodiment of the invention the multimodal welding power distribution is a tri-modal welding power distribution. In this case three beams are provided with one beam focussing on the trailing and the leading edge, and one beam in between these two beams focussing on the location where the edges abut. If the filler wire is introduced at the location of the middle beam, then the melting pool can be carefully controlled. By choosing suitable optics or a suitable laser configuration, this system also enables easy conversion from a single beam configuration to a twin-spot or a triple-spot configuration.

In an embodiment of the invention the multimodal welding power distribution is a bimodal welding power distribution.

This embodiment is the simplest embodiment in terms of number of spots with which one or more of the objects of the invention can be reached and is therefore a preferable embodiment for instance because of maintainability, limited complexity of the optical system and ease of operation.

The magnitude of generated defects due to the vertical and/or horizontal misalignments were found to be minimised by the twin spot process as more of the base metal was melted to form the joint. Larger horizontal and/or vertical misalignments can be tolerated and still obtain improved weld quality. The bimodal welding power distribution is the simplest form of welding power distribution with which the advantages according to the invention can be reached.

In some cases, it is beneficial to use a welding power distribution comprising a less severe thermal gradient away from the joint, for instance when welding material which is sensitive to cracking as a result of the high temperature gradient. By using a less steep welding power distribution towards the joint, for instance by introducing an additional peak on one or both sides of the joint, the temperature increase towards the joint can be made more gradual. However, care has to be taken to retain proper keyhole welding characteristics. In keyhole laser welding the laser beam produces a keyhole in the material by the presence of fast evaporating metal vapour or plasma. The keyhole moves with the focussed laser beam along the seam. The molten material flows around the keyhole from the front wall backwards and closes the gap. The presence of at least two foci in the bimodal welding power distribution, or more than two in the multimodal welding power distribution, results in more molten material between the foci.

In an embodiment the chromium-equivalent (Cr_eq (%)) and the nickel- equivalent (Ni_eq (%)) of the pool of molten material is delimited by the following equations:

Ni_eq (%) < 12% for Cr_eq < 8.4% (i) Ni_eq (%) > -2.5-Cr_eq + 9.5 (ii)

Ni_eq (%)> =+ 1.09-Cr_eq - 8,2 (iii)

Ni_eq (%) < -(Cr_eq/1.4) + 18 for Cr_eq between 8.4% and 14.5% (iv) and wherein

Cr_eq (%) = %Cr + %Mo + 1.5-%Si + 0.5-%Nb (v) Ni_eq (%) = %Ni + 30-%C + 0.5·%Μη (vi)

These equations delimit an irregular hexagon in a (Cr_eq, Ni_eq)-diagram and as long as the composition of the pool of molten material is such that the (Cr_eq, Ni_eq) of the material falls in this hexagon, then the mixture of molten filler wire and molten material of first and second strip results in combination in a bendable weld which is also cost effective. The inventors found that when the (Cr_eq, Ni_eq) composition of the pool of molten material has a Ni_eq above equations (ii) and (iii) (area M in figure 7a) also results in good properties, but this requires alloying of the melt pool with Ni and Cr to an extent that it is no longer and economical process. So the irregular hexagon delimits a technically and economically attractive process. In an embodiment of the invention the first metal strip and the second metal strip are provided in a coiled form.

In an embodiment of the invention, the welding power distribution is asymmetrical. In this embodiment the height and/or width of the first peak is different from that of the second peak, i.e. the welding power is different on both sides of the abutting edges to be welded together. This embodiment is particularly relevant for welding metal strips having different properties e.g. as a result of chemical composition, or for welding metal strips having a thickness difference.

In an embodiment of the invention the metal strips to be welded together are electrical steel strips and wherein the steel filler wire is a stainless steel filler wire.

The use of the stainless steel filler wire in combination with a laser beam welding process ensures a good quality weld and good properties of the weld . The most important property of a weld between strips or coils during processing in a continuous processing line is that it must stay intact. In the pickling line after the hot rolling mill the strip is repeatedly bended in a tension leveller within the production chain of pickled cold-rolled metal strips, scale breakers are typically installed at the entry of the pickling section to break the oxide layers on the hot rolled strip surface. Tension levellers are normally located at the exit of strip- processing lines, such as continuous galvanizing or finishing lines. The tension levellers improve strip flatness by reducing flatness deficiencies such as centre or edge buckles and strip camber, and minimize residual stresses of the finished strip. When running through such machines, the strip is bent alternately around multiple rolls with small diameters while high tension stresses are applied. The combined bending and tensile stresses may cause a weld to fail leading to extensive damage in the installations and large delays in production.

The use of a stainless steel filler wire in combination with a laser beam welding method provides a weld which is able to withstand the multiple bending operations and thereby enables stable production of electrical steels in continuous processing lines.

In a preferred embodiment of the invention the stainless steel filler wire is a wire of the 309 type such as, but not limited to, 309L. The inventors found that using this type of wire leads to a microstructure in the weld that is able to withstand the repeated bending.

In an embodiment of the invention the electrical steel is a non-grain oriented electrical (NGO) steel and has a silicon level of from 1 to 4.5% or the electrical steel is a grain oriented electrical steel (GOES) and has a silicon level of from 1 to 4.5%. In an embodiment of the invention the weld is subjected to a post welding heat treatment. The inventors found that subjecting the weld to a post welding heat treatment resulted in a further and significant improvement of the properties of the weld. The post welding heat treatment could be performed by means of heating means such as an induction coil or coils or low level laser beam(s).

The invention will now be further described by means of the following, non limitative examples under reference to the following, non-limitative figures.

Table 1 - Chemical composition (wt.%, except N (ppm)) of electrical steel (ES), low-carbon filler wires (SG2, V) and stainless steel filler wires.

Table 2 - Welding results Electrical Steel.

Table 2 clearly shows the beneficial effect of using a stainless steel filler wire on the bending properties instead of no wire (E) or instead of a SG2 wire, and also demonstrates the beneficial effect of the post welding heat treatment (PWHT). A result marked with is suitable for processing in a continuous processing line. The bending properties were determined using a test set-up in which welded samples are clamped and repeatedly bent over an angle of +45° and -45° until the sample fails. The bending takes place at a radius of 60 mm, which results in a stress in the surface comparable to a tension levelling device in a production line. The value until failure was determined on the basis of three samples and averaged in Table 2.

Figure 1 shows the hardness (HV) as a function of the distance (mm) over the weld for specimen H (diamonds, no PWHT) and I (squares, PWHT). Clearly the hardness of the weld is much higher than that of the base material. The microstructure of the weld is also quite different from that of the base material, as can be clearly observed from the figure 2 (H), 3 (I), 4 (H) and 5 (I) which show the weld and a close up. The inventors believe that the improvement in bending properties by the PWHT is caused by recrystallisation of the base material around the weld such that polygonal equi-axed grains extend through the sheet thickness at the weld. These grains are relatively large and are therefore easily deformable. The absence of internal stresses may also contribute to the favourable bending properties.

Figure 6 shows a schematic representation of a symmetrical bimodal welding power distribution (Figure 6a) and an example of a asymmetrical bimodal welding power distribution (Figure 6b) with a first power peak (A) and a second power peak (B). The welding power is schematically plotted on the vertical axis. The spatial separation between the first power peak and a second power peak is indicated with d. The location of the trough is indicated with C. The dashed line indicates the axis along which a beam from a single spot laser would target the plane where the edges (2, 4) abut. The length of the first and the second metal strip is obviously not to scale, but these extends beyond the boundaries of the figure. Only the end parts of the first and the second metal strip are shown in Figure 6a-c. The abutting edges, i.e. the location where the two metal strip or coil ends 2 and 4 are to be connected to each other by welding, is indicated with 5. In the figures the welding power per spot is schematically indicated as a Gaussian type curve. Symmetrical bimodal welding power distributions are preferable in welding materials of similar welding behaviour, such as similar materials with similar thickness. The asymmetrical bimodal welding power distribution can be used in welding materials which have dissimilar welding behaviour, such as similar materials of different thickness, or different materials of similar or dissimilar thickness. An asymmetrical bimodal welding power distribution can generally be achieved by reducing the welding power to one of the laser beams, for instance with a diaphragm. Figure 6c shows an asymmetrical multimodal welding power distribution with a first peak (A) separated from a second peak (B) with a separation d, and a third peak (C) focussing on the abutting edges. The dashed curve shows a symmetrical multimodal welding power distribution.

Figure 7a and b show the irregular hexagon in the (Cr_eq, Ni_eq) diagram. The area of the diagram delimited by the x and y-axis and the equations (ii) and (iii) delimits an area (marked with M) suitable for producing welds that have good bending properties and good hardness values. Figure 7b shows an irregular hexagon (marked with M) delimited by the x and y-axis and the equations (i) to and (iv) which is an area suitable for producing welds that perform well during bending and have good hardness values, as well as an economically attractive low alloying content of the melt pool. This allows using lower cost filler wire, than the area marked with M in figure 7a. If the (Cr_eq, Ni_eq) lies in the area marked with F, then this means that ferrite will be present in the solidified weld, and this results in inadequate properties.

It is clear that the (Cr_eq, Ni_eq) position of the molten mixture of filler wire and strip material depends on the ratio of filler wire to strip material. This is indicated in the legend by 13, 15 and 25% of filler wire and the remainder is NGO. The amount of filler wire depends on the welding speed and the speed of feeding wire. E.g. by feeding wire of diameter 1 mm at a speed of 1.6 m/min and a welding speed of 1.6 m/min a length of 1 mm of weld pool is provided with π/4 mm³ filler wire. By varying the welding speed and power of the laser, the size of the weld pool is determined. The size of the weld pool can be easily determined from the cross section of the weld, and therefore a relation between weld pool volume and welding parameters can be established. Knowing the size of the volume of the pool the ratio of filler wire and strip material can be determined and this allows the check whether the (Cr_eq, Ni_eq) position lies above lines (ii) and (iii) or within the irregular hexagon respectively.

Figure 9a-h depict the JMatPro^® (Public Release Version 6.1) results of the equilibrium phases as a function of the wire, the material to be welded and the ratio between material to be welded and the filler wire in the melt pool (87-13 mix means 13% filler wire). This mix is determined as described above, and this mix determines the composition of the melt pool which is entered into JMatPro. The liquid phase is indicated with L, ferrite with a and austenite with y. The low temperature ferrite in figure 9e to h will transform to martensite thus resulting in the desired combination of 100% austenite and a martensitic final structure (note 9c also results in martensite, but this is 100% wire). The graphs in figure 9e to h indicate ferrite to be present instead of martensite, but this is because these figures are equilibrium graphs. Therefore "a" has to be interpreted as martensite.

Figure 9a indicates that NGO is purely ferritic, and can therefore not transform to martensite, because there is no austenite present to produce the martensite upon cooling at any temperature. Figure 9b shows that GOES material does not become 100% austenite, and therefore does not satisfy the requirement for a fully austenitic microstructure when welded without a filler wire even though the (Cr_eq, Ni_eq) of the GOES composition falls in the hexagon of figure 7. Figure 9d shows the weld pool composition using a SG2 filler wire showing that the material does not become 100% austenite even when introducing 24% filler wire in the weld pool. Figures 9e to h show a weld pool composition which becomes fully austenitic upon cooling at 13% filler wire in the mixture.

Figure 8 shows the bending (# = 1 means one bending to and fro) and formability (Erichsen test) of these combinations by twin-spot welding (YAG, 0.6 mm mirror). A number above 20 is generally considered to be excellent. Figure 8a shows that the formability of a weld increases if there is a gap between the first steel and the second steel. When looking at the cross section itself the inventors believe that this may be due to the absence of the bulge in case of the gap. The Erichsen test is performed by pressing a punch into the weld until the weld or the material breaks or tears. The distance or height that the punch has travelled is expressed in millimetres. The measurement is stopped after rupture of the weld. The Erichsen cupping tests were performed according to ISO standard 20482 :2003. Figure 8b and c show the number of to-and-fro bending operations before cracks appear. During this test, the material including the weld is bent with a radius of 60 mm. This results in a stress comparable to the surface strain that is applied by the tension leveller (TRM) of a pickle line. Next the material is re-bent in the opposite direction and is repeated until it fractures. The results are expressed in the number of to-and-fro cycles. Figure 8b and c show that the used wires all provide good welds with good properties. 316 gives the best performance and is actually one of the cheapest wires of those investigated. Figure 8c shows that welding NGO without a filler wire is not an option. Similar results are obtained for welding GOES- strips.

Claims

Method of producing a steel strip by laser welding the trailing edge (2) of a first steel strip (1) to the leading edge (4) of a second steel strip (3) wherein a filler wire is used and wherein the laser welding is performed by a laser that provides

(i) . a single spot power distribution, or

(ii) . a multimodal welding power distribution, wherein the multimodal welding power distribution has at least a first power peak (A) and a second power peak (B) separated from the first power peak by a power trough (C), wherein the first power peak converges on the trailing end of the first steel strip and wherein the second power peak converges on the leading end of the second steel strip,

wherein at least one of the steel strips to be welded together is an electrical steel strip;

wherein the pool of molten material created during welding has a composition comprising a mixture of the molten filler wire and the first and second steel strips;

wherein the composition of the pool of molten material is such that after solidification of the pool of molten material, the solidified material becomes a fully austenitic microstructure at a point in time during cooling of the solidified material and that the said fully austenitic microstructure of the solidified material subsequently is transformed into a martensitic microstructure.

Method according to claim 1 wherein the chromium-equivalent (Cr_eq (%)) and the nickel-equivalent (Ni_eq (%)) of the melt pool is delimited by the following equations:

Ni_eq (%) < 12% for Cr_eq < 8.4% (i) Ni_eq (%) > -2.5-Cr_eq + 9.5 (ii) Ni_eq (%)> =+ 1.09-Cr_eq - 8,2 (iii) Ni_eq (%) < -(Cr_eq/1.4) + 18 for Cr_eq between 8.4% and 14.5% (iv) and wherein

Cr_eq (%) = %Cr + %Mo + 1.5-%Si + 0.5-%Nb (v) Ni_eq (%) = %Ni + 30-%C + 0.5·%Μη. (vi)

Method according to claim 1 or 2 wherein the martensitic microstructure of the solidified material contains at least 90% martensite, preferably at least 95% martensite, more preferably only martensite.

Method according to any one of claims 1 to 3 wherein the first and the second steel strip to be welded together are electrical steel strips.

Method according to any one of claims 1 to 4 wherein the carbon content of the martensite is at most 0.4%.

Method according to any one of claims 1 to 5 wherein the filler wire is a stainless steel wire of the 309 or 316 type or a filler wire comprising at least 20% Cr and at least 10% Ni.

Method according to any one of claims 1 to 6 wherein the filler wire is a stainless steel wire of the 309L type.

Method according to any one of claims 1 to 7 wherein the electrical steel is a non-grain oriented electrical (NGO) steel and has a silicon level of from 1 to 4.5%.

Method according to any one of claims 1 to 8 wherein the electrical steel is a grain oriented electrical steel (GOES) and has a silicon level of from 1 to 4.5%.

Method according to any one of claims 1 to 9 wherein the multimodal welding power distribution is a bimodal welding power distribution.

Method according to any one of claims 1 to 10 wherein the first steel strip and the second steel strip are provided in a coiled form.

Method according to any one of claims 1 to 11 wherein during welding of the first steel strip to the second steel strip there is a gap between the leading edge of the second steel strip and the trailing edge of the first steel strip wherein the gap is between 0.1 and 1 mm, preferably between 0.15 and 0.5 mm.

Method according to any one of claims 1 to 12 wherein the weld is subjected to a post welding heat treatment.