METHOD FOR MANUFACTURING A WELDED ASSEMBLY

FIELD OF THE INVENTION

The present invention relates to methods for manufacturing a welded assembly comprised of two or more sections of metal stock. The method comprises the step of welding the sections of stock together, and the subsequent step of heat treating the welded assembly in order to normalize the properties of the heat-affected areas with remaining portions of the welded assembly.

BACKGROUND OF THE INVENTION

In the manufacture of metal sheet, strip or plate products, there are limitations imposed on the manufacturer in terms of the dimensions and other properties which can be imparted to such products. For example, conventional casting and rolling mills have limitations in the thickness and width of material which may be produced. Furthermore, each piece of stock produced by the mill, whether in the form of coiled or flat stock, is typically manufactured to have constant thickness, composition and metallurgical properties throughout.

On the other hand, industrial users of these products frequently require stock which exceeds the width parameters of the mill, or which is of variable thickness, composition or metallurgical properties. A number of solutions have been proposed for addressing this problem. Many of these solutions involve the joining together of two or more pieces of stock by the end user of the product. In cases where the stock is of relatively thin gauge, such as for automotive applications, two or more pieces of stock may be butt welded by a laser beam. An example of this type of process is disclosed in U.S. Patent No. 6,034,347 (Alber et al.) which involves laser butt welding of metal sheets for car body building. Another example is disclosed in U.S. Patent No. 5,250,783 (Nishi et al.) which involves laser butt welding of metal sheets of different thicknesses. Yet another example is disclosed in U.S.

Patent No. 6,642,474 (DeIIe Piane et al.) which involves laser butt welding of sheet blanks of different thicknesses and/or different materials.

While laser welding is suitable for joining relatively thin sheet materials, it is not suitable for joining materials having a thickness greater than about 2.5 mm, such as metal plate stock. The use of lasers in thicker materials may result in a non-uniform V-shaped weld joint resulting from greater melt pool formation on the side facing the laser. The solutions proposed for overcoming these problems are not entirely satisfactory. For example, U.S. Patent No. 4,223,201 (Peters et al.) discloses two-sided laser butt welding of metal plates for the fabrication of shell sections for ship construction; and U.S. Patent No. 5,994,665 (Nishibayashi et al.) discloses an initial butt welding operation by a laser or plasma arc which joins a plurality of moving sheets together, followed by pressure welding in a finishing mill. The Peters et al. process is complex and costly due to the need for two laser welding apparatus, and in Nishibayashi et al. the width of the final product is limited by the width limitations of the mill.

Electron beam welding is known to be effective for joining together relatively thick articles, for example steel articles having a thickness of up to about 200mm. An example of a laser butt welding process is disclosed in U.S. Patent No. 6,888,090 (Murphy), which relates to electron beam butt welding of dissimilar metal subcomponents having a thickness of 100mm or more.

Conventional welding methods are typically carried out on stock which has already been heat treated. The heat treatment provides the stock with desired properties of hardness, strength, toughness, etc. Subsequent welding of the stock may alter the chemical composition and/or the metallurgical microstructure along the weld joint and in adjacent areas of the stock which are heated to high temperature, thereby altering or limiting the performance of the welded assembly. This problem has been addressed in the prior art by providing a thermal pre- or post-treatment along the weld joint in order to reduce the loss in properties brought about by welding. An

example of such a process is disclosed by US 2004/0188394 Al (Becker et al.)- This process utilizes a single laser beam to carry out both the welding and thermal pre- and post-treatment steps in order to reduce the loss in ductility brought about during welding of high stiffness steels.

There remains a need for improved methods for manufacturing welded assemblies, and particularly for such methods in which the weld joint is normalized with the components being joined.

SUMMARY OF THE INVENTION

According to one aspect, the invention provides a method for manufacturing a welded assembly. The method comprises: (a) providing a first component comprised of a first metal, the first component having a first surface, an opposed second surface and an end surface joining the first and second surfaces and defining a thickness of the first component; (b) providing a second component comprised of a second metal, the second component having a first surface, a second surface and an end surface joining the first and second surfaces and defining a thickness of the second component; (c) aligning the first and second components so that they are in contact with one another along an abutment line; (d) welding the first and second components together to form an as-welded assembly in which the first and second components are joined together by a weld joint, wherein the weld joint extends along said abutment line; and (e) after formation of the weld joint, heat treating said as-welded assembly by first heating said as-welded assembly to a pre-determined temperature and then cooling it.

In another aspect, the invention provides a welded assembly produced by the method of the invention, wherein said first and second components are produced by a casting and rolling mill and have a width dimension which is less than a maximum width dimension which can be produced by the casting and rolling mill; and wherein said welded assembly comprises a plate having

a width dimension greater than said maximum width dimension and having a substantially homogeneous or tailored microstructure and properties.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will now be described, by way of example only, with reference to the accompanying drawings in which:

Figure 1 schematically illustrates portions of a pair of components to be joined according to the method of the invention;

Figure 2a schematically illustrates the components of Figure 1 after they have been aligned with their end surfaces in abutment, wherein the components are of the same thickness;

Figure 2b schematically illustrates the components of Figure 1 after they have been aligned with their end surfaces in abutment, wherein the components are of different thickness;

Figure 3 is a schematic side view of the components of Figure 1 being welded according to the method of the invention;

Figure 4 is a schematic perspective view of the components of Figure 1 being welded according to the method of the invention;

Figure 5 is a schematic perspective view of the components of Figure 1 after the welding is completed;

Figure 6 schematically illustrates a welded assembly according to the invention, following a heat treatment step;

Figure 7 is a representative set of photographs/radiographs showing the weld produced in run 21;

Figure 8 is a photograph showing a transverse section through the weld joint of run 21; and

Figure 9 is a photograph showing a transverse section through a weld joint before and after heat treatment.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

Figure 1 illustrates a first component 10 and a second component 12 which are to be joined together according to a preferred method of the present invention. The first component 10 is comprised of a first metal and has a first surface 14, an opposed second surface 16 and an end surface 18 joining the first and second surfaces 14, 16. The end surface 18 has a thickness Tl and a length Ll. Similarly, the second component 12 is comprised of a second metal and has a first surface 20, an opposed second surface 22 and an end surface 24 joining the first and second surfaces 20, 22. The end surface 24 has a thickness T2 and a length L2.

In the embodiment of Figure 1, the end surfaces 18, 24 are rectilinear, meaning that they comprise planar surfaces which are substantially perpendicular to the first and second surfaces of components 10, 12. It will, however, be appreciated that end surfaces are not necessarily rectilinear, i.e. they may be of any desired geometry which permits the formation of a weld joint. For example, the end surfaces may be bevelled along their thickness dimension, they may be curved either along their length dimension and/or their thickness dimension, and they do not need to be substantially perpendicular to the first and second surfaces. For example, the end surfaces can be arcuate along the thickness dimension as shown in Figure 4 of the above-mentioned patent to Murphy. However, for ease of welding, it may be preferred that the end surfaces 18, 24 are rectilinear.

The thicknesses Tl and T2 of the components 10, 12 are preferably up to about 200 mm, more preferably from about 2.5 to about 200 mm, and even more preferably from about 3 to about 100 mm. It will be appreciated, however, that the method of the present invention is applicable to a wide range of gauges, ranging from light gauges in the range of 6 mm or less to thick gauges up to about 200 mm in thickness. The thicknesses Tl and T2 may be the same or different. In Figure 1, the components 10, 12 are shown as being of constant thickness in the regions of end surfaces 18, 24. It will be appreciated that this is not necessary, and that the components 10, 12 may be of variable thickness in these regions. Furthermore, it will be appreciated that the components 10, 12 may be of variable shape and thickness outside the regions adjacent to the end surfaces 18, 24, and are not necessarily flat. For example, components 10, 12 may preferably comprise shaped articles.

The components 10, 12 are preferably comprised of steel, but may instead be comprised of other ferrous or non-ferrous metals. Furthermore, the metal of the first component 10 may be either the same as or different from the metal of the second component 12.

In a particularly preferred embodiment of the present invention, the components 10, 12 are in a "green" state. As used herein, the term "green" means that that the components 10, 12 have been rolled to their final thickness dimensions in a hot rolling mill and are intended to be heat treated in a subsequent processing step, but have not yet undergone the heat treatment. The as-rolled green material is preferably allowed to cool prior to welding, and is preferably cooled to ambient temperature. The microstructure of the green material is irrelevant since the green microstructure is destroyed during the subsequent heat treating step.

In some embodiments of the invention, the components 10, 12 are in the form of metal sheet, strip or plate, and the method of the invention involves the formation of a butt joint between the components 10, 12. However, it

will be appreciated that the components can be of various shapes and sizes, and that the joint between the components 10, 12 is not necessarily a butt joint. For example, the first and second components 10, 12 can be lengths of pipe which are to be butt welded together in end-to-end fashion, in which case the end surfaces 18, 24 would be annular. Alternatively, the components 10, 12 may form part of the same article, as in the case where the article is a pipe which is to be sealed by a longitudinal weld seam. Accordingly, it will be seen that the components 10, 12 can have a variety of shapes and sizes. Furthermore, it will be appreciated that the end surfaces 18, 24 do not necessarily abut one another. Rather, they may overlap one another so as to form a lap joint. For simplicity, the following description will focus on embodiments of the invention where both components are in the form of sheets, strips or plates and where a butt joint is formed between the components.

As shown in Figure 2a, the components 10, 12 are brought together and aligned with their respective end surfaces 18, 24 in abutment with one another along an abutment line 26. In the case where the end surfaces 18, 24 are planar, the abutment line 26 lies in an abutment plane 28 which comprises a contact surface interface between the end surfaces 18, 24. In Figure 2a, the abutment plane 28 has an area which is equal to the area of end surface 18 (Tl x Ll) or end surface 24 (T2 x L2).

Figure 2b illustrates a situation where two components 10, 12' of different thicknesses are brought together. Component 12' has a thickness T2' which is less than the thickness Tl of component 10. As in the embodiment shown in Figure 2a, the end surfaces 18 and 24' (not shown) of components 10, 12' are in abutment with one another along an abutment line 26'. The abutment line 26' lies in an abutment plane 28' which comprises a contact surface interface between the end surfaces 18, 24'. In Figure 2a, the abutment plane 28' has an area which is equal to the area of end surface 24' (T2' x L2').

In the next step of the method, the components 10, 12 are welded together along the abutment line 26. In the embodiment of the invention described below, the components 10, 12 are comprised of relatively thick plate stock and are joined by electron beam welding. It will, however, be appreciated that welding can instead be performed by any other type of suitable welding apparatus, including laser welding, plasma arc welding, submerged arc welding, gas metal arc welding, inert gas welding, etc. For example, a laser welding apparatus may be suitable for use in the present invention, particularly where the materials are in the form of relatively thin metal sheets, typically not exceeding a thickness of about 2.5 mm.

The electron beam welding apparatus includes an electron gun 30 which generates a beam of high-energy electrons 32 and may further comprise a high voltage power source, a vacuum chamber or enclosure, pumping equipment, a workpiece or gun manipulator and a control system, none of which are shown or described here. The electron gun 30 has a thermal cathode from which the electrons are accelerated by a high potential in the range from 30-200 kV, preferably in the range from 100-150 kV, and having a power of up to about 100 kW. The welding apparatus preferably also includes means for magnetically directing and focusing the electron beam 32 in close proximity to the abutment line 26. The electron beam 32 may be generated in a vacuum of at least 5 x 10 ^"5 mbar, although the components 10, 12 may be housed in a chamber maintained at a coarser vacuum level such as 5 xlO ^"3 to 10 mbar. Alternatively, in the case of Non-Vacuum Electron Beam (NVEB) welding, the components 10, 12 may not be housed in a vacuum, so that the electron beam 32 traverses the atmosphere between the gun 30 and the components 10, 12. In yet another embodiment, the electron gun 30 may preferably form part of a reduced pressure electron beam (RPEB) welding apparatus in which the pressure in the welding enclosure is from about 10 ^"1 - 10 mbar.

The electron beam 32 is directed at components 10, 12 in sufficiently close proximity to the abutment line 26 so as to form the weld joint 34 at the abutment line 26. In the embodiment shown in Figure 3, the electron beam

30 is directed at the abutment line 26, but it will be appreciated that the electron beam 30 may instead be directed at a point which is spaced from the abutment line 26 and/or may have a transverse component to its movement, for example as disclosed in Becker et al. The electron beam 30 is preferably parallel to the abutment plane 28 may preferably be coplanar with the abutment plane 28.

The power of the electron beam 32 is preferably sufficient so as to completely penetrate through the components 10, 12 and to form the weld joint 34 in a single pass. The electron beam 32 causes localized melting and vaporization of the components 10, 12 in the region of the abutment line 26, thereby forming weld joint 34 which joins the components 10, 12 together. Where, as shown in Figure 3, both components 10, 12 have the same thickness, the electron beam 32 must penetrate completely through the thickness of the components 10, 12. In cases where one of the components is reduced in thickness, the electron beam 32 may only need enough power to penetrate through the thinner of the two components. Where the welding is performed by a laser beam, it preferably also has sufficient power so as to penetrate the components 10, 12 and form a weld joint in a single pass.

As the electron beam 32 is being directed at the components 10, 12, it is advanced along the abutment line 26 so as to extend the weld joint 34 continuously along the abutment line 26. Figure 4 illustrates the electron beam 32 after it has been advanced part way along the abutment line 26. It will be appreciated that the electron beam 32 may be advanced along line 26 either by moving the electron gun 30 along the abutment line 26 in the direction of arrow A, by moving the components 10, 12 along the abutment line in the direction of arrow B, or both.

Figure 5 illustrates the components 10, 12 after they have been joined by the weld joint 34. As shown, the weld joint 34 extends continuously along the entire length of the abutment line (i.e. lengths Ll and L2) and extends through the thickness of the components 10, 12 throughout its entire length.

Therefore, the components 10, 12 are welded together throughout the entire abutment plane 28 to form an assembly 36. This assembly 36, which has not yet undergone heat treatment, is referred to herein as the "as-welded assembly" and is identified by reference numeral 36.

The welding process elevates the temperature of the components 10, 12 in the region of abutment line 26 to a temperature greater than the melting points of the materials comprising components 10, 12, thereby forming a molten pool of material which is allowed to cool and freeze to form a weld joint 34, which comprises a bead or seam along the abutment line 26. The rapid heating and cooling of the material in the region of abutment line 26 eliminates the original metallurgical microstructure of the materials comprising the as-rolled components 10, 12.

The welding is preferably carried out without the introduction of weld metal from an external source such as a consumable welding rod or the like. In the alternative, an external source of weld metal may be used where the external weld metal has a composition and properties which are similar to or compatible with the composition and properties of the components 10, 12 to be joined.

The region immediately adjacent to the weld joint 34 does not melt and refreeze but is subject to temperatures up to the melting temperature and is generally referred to as the Heat Affected Zone (HAZ) 35. The approximate boundaries of the HAZ 35 are indicated by dashed lines in Fig. 3. The metallurgical microstructure in the HAZ is also altered by exposure to heat from the welding operation followed by cooling, with the relatively cooler portions of components 10, 12 acting as a heat sink. As a result, the as- welded assembly 36 has a variable metallurgical microstructure across the weld joint 34. In addition, a complex three-dimensional residual stress distribution is set up throughout the weld joint 34. Where the components 10, 12 are plates, this can affect flatness of the as-welded assembly 36.

In the method of the present invention, the as-welded assembly 36 is heat treated such that the weld joint 34 is normalized with the remainder of the material comprising the as-welded assembly 36. Once the assembly 36 undergoes heat treatment, it is referred to herein simply as the "welded assembly", also identified by reference numeral 36. The heat treatment preferably performed by heat-treating the entire as-welded assembly 36 in one or more steps, depending on the desired properties of the final welded assembly 36.

In the case where components 10, 12 are comprised of the same metal, the heat treatment provides the entire welded assembly 36 with a substantially homogeneous microstructure, thereby ensuring that the welded assembly 36 has substantially uniform physical properties throughout. In the case where the components 10, 12 are comprised of metals having different chemical compositions, the heat treatment provides a transition zone extending across the weld joint 34 in which there is a gradual transition between the metal comprising component 10 and the metal comprising component 12.

Preferably, the heat treatment includes a step in which the entire as-welded assembly 36 is heated to a pre-determined temperature, followed by cooling. Preferably, the entire as-welded assembly 36 is heated in a furnace, and the pre-determined temperature is sufficiently high to bring about a transformation in the metallurgical microstructure of the as-welded assembly 36. For example, where the components 10, 12 are comprised of steel, the as-welded assembly 36 is heated to a temperature, and for an amount of time, sufficient to convert the steel microstructure from ferrite to austenite. Typically this temperature is from about 800-1000 ⁰ C, with the exact temperature and time of heating being dependent on a number of factors including carbon content and thickness of the assembly 36. The temperature is typically about 50 ⁰ C above the temperature range for conversion from ferrite to austenite at the carbon content of the steel comprising the assembly 36.

Once a fully austenitic state is reached, the assembly 36 is cooled to ambient temperature either by cooling in air or more rapid quenching with water. Quenching with water is preferred and involves spraying the austenitic steel with water jets so as to very quickly cool it to ambient temperature, thereby freezing the steel and typically resulting in transformation of the austenite to a relatively hard martensitic phase.

In some preferred embodiments of the invention, the heat treatment may comprise a subsequent tempering step. For example, the assembly 36 may be heat treated and cooled or quenched to ambient temperature as described above, and then re-heated to a pre-determined temperature which is lower than the temperature of the first heat treatment for a sufficient amount of time to temper the steel. Following the second heating step, the welded assembly 36 is once again cooled or quenched to ambient temperature. The pre-determined temperature of the second heat treatment is below the temperature range at which the steel microstructure is converted to austenite, preferably in the range from about 200 ⁰ C to less than 800 ⁰ C, for example from about 200-735 ⁰ C, typically about 200-700 ⁰ C and more typically about 200-300 ⁰ C . As mentioned above, this second heat treatment step tempers the steel, refining the microstructure of the article 36 and allowing carbon to diffuse away from the hardened martensitic regions. This has the effects of reducing brittleness and improving properties such as toughness and malleability.

The method of the invention may also include the step of evaluating the physical properties of the assembly in the region of the weld joint and comparing these physical properties with those of the remaining portions of the assembly, or with welding processes in which the components to be joined are heat treated prior to welding. One or more of the following tests may be performed:

(a) tensile testing perpendicularly across the weld joint to determine a stress strain curve from which are derived yield strength, ultimate tensile strength and elongation;

(b) Charpy impact testing of the weld joint and HAZ;

(c) bend testing across the weld;

(d) microstructural evaluation including qualitative assessment of microstructural constituents, morphologies, and any defects such as cracks, inclusions or porosity;

(e) micro hardness through the thickness and through the weld zone; and

(f) visual assessment of the weld region including general appearance of weld bead and adjacent areas, and surface characteristics including surface scale, roughness or pits or areas below nominal thickness. The welded assembly 36 produced by the present invention may be in the form of a shaped article, for example where the components 10, 12 have been pre-shaped. More typically, the welded assembly 36 is in the form of a plate or blank, for example where the components 10, 12 are in the form of flat strip, sheet or plate stock. Where the welded assembly 36 is in the form of a plate or blank, it may be subjected to shaping by the end user, depending on the intended use of the assembly.

Where the components 10, 12 comprise strip, sheet or plate stock of different thicknesses and/or compositions, the resulting welded assembly 36 may comprise a "tailor-welded" blank. These products are widely used in the automotive industry, for example in the manufacture of automobile body parts.

As will be apparent from the above description, the final width of the welded assembly 36 is not limited by the width capacity of the casting and rolling mill. The present invention allows the production of relatively wide blanks with homogeneous or tailored composition and properties. These blanks are useful in the manufacture of a range of products requiring plate material with large dimensions, including refuse and other service trucks; armour for civilian and military vehicles and for stationary applications; mining and construction equipment; rail cars and ships.

The potential benefits of the present invention includes the opportunity to reduce the number of parts by amalgamating welded assemblies, optimizing the composite blank for weight and function, reduce welding in manufacturing the part, reduced design and development time, potential lower manufacturing costs, weight reduction, improved structural performance and improved utilization of material. The invention may be targeted at applications where the part or assembly performance requirements vary in different locations, where the part size is too large to be supplied from a single piece of material. Examples of applications embodying the invention include the side-walls of refuse vehicles, containers, utility boxes or specialty vehicles; dump bodies; wear- resistant liners for equipment, vehicles, dump and other utility bodies; conveyor systems; ballistic doors and shields; vehicle and equipment chassis.

The invention will now be further illustrated by the following Examples.

EXAMPLES

Laboratory scale Non-vacuum Electron Beam (NVEB) welding trials were carried out to establish conditions for producing welds of appropriate quality and profile in a number of steel samples representative of a variety of types and thicknesses of steel, without the use of filler wire. The trials were carried out initially by producing melt runs, and the best conditions were repeated on samples prepared as butt joints. The results showed some variability in both weld bead profile and surface profile. This variability was associated with a variety of factors including plate condition, joint fit-up, NVEB process gas shielding and welding process details.

The trial demonstrates that NVEB welding of a variety of types and thicknesses of steels is capable of producing welds that meet the stringent quality requirements of ISO 13919-1 (Welding - Electron and Laser-Beam Welded Joints - Guidance on Quality Levels for Imperfections - Part 1 : Steel). Weld bead quality was influenced to some extent by joint preparation details and some welding process controls. It is anticipated that

these issues can be addressed by engineering solutions and some additional process controls during scale-up in a mill setting.

Initial tests were carried out to examine the feasibility of using Reduced Pressure Electron Beam (RPEB) welding. The tests indicated that RPEB is feasible, but that better joint fit-up and preparation were required than were achievable by shearing in order to produce acceptable weld profiles. For this reason the trials were conducted using NVEB.

Another set of initial tests were conducted on plate using NVEB welding in the flat or down-hand welding position (ASME, IG) to establish basic feasibility. While these weld samples were found to be sound by radiographic testing and close to satisfactory in terms of weld profile and reinforcement, it was decided to conduct the remaining trials with welding performed in the horizontal-vertical position only, i.e. with the beam axis horizontal, the plate standing vertical and traversing horizontally. This is designated as the 2G position in the ASME IX code and as the PC position in the BS EN ISO standard. This orientation was selected for the reason that it is more practical for application in a mill setting and is expected to produce a better quality weld profile under mill conditions.

The possibility of employing a pre-treatment using a flux to enhance weld bead formation was discussed and briefly tested using titanium oxide. As a result of these tests, it is believed that the use of flux could form the basis of a method for weld bead control.

The examples described below were conducted to establish NVEB welding conditions which provide butt joints giving fully proud weld beads in various types and thicknesses of steels.

A - Materials

The materials used in the examples are described in Table 1 below.

Table 1 - Materials

B - Welding Equipment and Conditions

The NVEB welding equipment used in these experiments consisted of a prototype high power gun column and traverse assembly. The electron source was RF excited, indirectly heated, diode gun. The accelerating voltage used in all trials was 150 kV, the beam current level ranged from about 100 - 180 mA, and the focus ranged from 1.7 to 1.8 Angstroms. The traverse was capable of speeds ranging from 200 to 4500 mm/min. All the welding was conducted in the 2G/PC position (beam axis horizontal, plate standing vertically and traversing horizontally). The weld specimen jig mounted on the traverse incorporated a rear gas shield enclosure and an excess beam capture plate. Argon gas was fed to the enclosure, usually at a flow rate of 10 l/min with an over-pressure level (measured at the gas bottle) of 2 bar.

The gun column was equipped with an annular wire brush surrounding the extremity of the column; this was in close contact with the weld specimen during welding to avoid ingress of air. Helium shielding gas was injected into the output nozzle of the gun column, flooding the wire brush shield region.

The welding was generally conducted at a working distance of less than 25 mm, and more typically from about 13 to 20 mm to avoid excess beam spreading.

The welding trials were carried out on five thicknesses of steel plate, ranging from 5 to 18 mm. Beam power and welding speed, the main weld penetration and profile control parameters, were dictated by plate thickness.

The samples which underwent the welding trials were examined by X- radiography, and photographs were made of the weld beads of some of the welded samples.

Example 1 - 6.4 mm Algoma Steel

Melt run trials on the 6.4 mm Algoma steel (5E5306) were conducted, and welds were subsequently made. Welding speeds ranged from 2000 to 4500 mm/min and beam current levels ranged from 100 to 165 mA. The melt run trials established that a speed of 3500 mm/min and a beam current of 135 mA were preferred. The beam current was raised to the range of 160 - 165 mA in some of the welding trials to ensure more adequate penetration and fuller beads.

In terms of joint preparation, the joint gap varied from 0 to 0.5 mm. The weld beads were typically undercut where the joint gap was the greatest, and were generally acceptable, ranging from proud to flat, where the joint gap was less than 0.15 mm. Plate mismatch was also an issue in some of the samples.

During the trial, the effect of shielding gas was also explored. Some runs were conducted without gas shields at the front and/or back of the joint. As would be expected, the beads of unshielded samples were oxidized more than with gas shielding, but the bead profile was still acceptable.

Run nos. 1-9 and 19-21 in Table 2 below were conducted with 6.4 mm Algoma steel.

Example 2 - 5 mm Algoma Steel and 5.8 mm EH36 Ship Plate

Melt run trials were conducted with the EH36 Ship Plate (5E53337) at a welding speed of 3500 mm/min and a helium over-pressure of 3 bar. Although penetration could be achieved at 130 mA, the beam current was increased to 150 - 155 mA for some of the welding trials in order to produce more fully proud weld beads. There welds showed some minor undercut on the back bead, but the front was fairly flat, with the weld beads ranging in width from about 2.8 - 4.5 mm. Weld trials were performed for both the EH36 Ship Plate and the 5 mm Algoma Steel.

Run nos. 14 and 15 in Table 2 below were conducted with 5 mm Algoma steel and run nos. 10-13 were conducted with EH36 Ship Plate.

Example 3 - 9.5 mm Steel

Lower welding speeds were used for the 9.5 mm steel samples (5E5335), ranging from 1000 - 2500 mm/min in order to allow welds to be made at the maximum beam power range available. At 1000 mm/min, full penetration was achieved even at 100 - 120 mA but the beads were quite wide. The beam current was increased to 153 mA for some samples to achieve full penetration at 2000 - 2500 mm/min, producing beads which were either proud or undercut to a minor extent, the beads ranging in width from 3.8 - 4.2 mm.

Run nos. 16-18 in Table 2 below were conducted with 9.5 mm steel.

Example 4 - 18 mm steel

The 18 mm steel (5E5338) has low C (0.06%) and high manganese (1.77%) with proven weldability in RPEB tests. With a beam current level of 170 mA, the welding speed was varied from 800 - 1500 mm/min, providing beads of variable profile. A welding speed of 800 mm/min surprisingly provided the best bead profiles, with the front bead exhibiting 0 - 3 mm of sink and the back bead being generally proud but deeply undercut in several places.

Run no. 22 in Table 2 below was conducted with 18 mm steel.

Conclusions - Welding Trials

The trials show that good preparation and presentation of the joints for welding are important. In these laboratory trials it was found that poor joint fit-up and flatness tolerances contributed to variability in weld appearance. In the production situation, it is believed that the issues of plate flatness, out of plane distortion and fit-up could be addressed by heavy engineering typical of secondary steelmaking practice.

The acceptance criteria in the BS EN ISO guidelines (ISO 13010-1 Welding) allow some undercut or underfill in moderate (0.15t or 1 mm), intermediate (O.lt or 0.5 mm) and stringent (0.05t or 0.5 mm) quality levels. A number of the welds produced in this trial could met the intermediate or even the stringent quality levels, particularly in the 5 and 6.4 mm thick Algoma Green 100 steel samples.

The welding speed in the trials was limited to 4500 mm/min and, at these speeds, the power levels used were sufficient for welding the materials of 5 - 9.5 mm thickness. However, the welding of the thicker 18 mm material would have benefited from better beam quality, higher power levels and/or pulsed operation with greater average power. It will be appreciated, however, that the above trials were performed on relatively thin materials primarily for reasons of practicality, i.e. it would have been relatively impractical to ship very thick plate samples for testing. It is well known that electron beam welding is relatively insensitive to the thickness of material and therefore it is expected that the process according to the present invention is applicable to materials of much greater thickness, at least up to about 200 mm.

Table 2 - Run. Nos. 1-22

Where the run no. in Table 2 above includes the letter "M", it is a melt trial, and where the run no. includes the letter "W" it is a weld trial. Under "Conditions", the column headed with "mA" shows the beam current level in mA, the column headed with "Speed" shows the travel speed of the sample in mm/min, and the column headed "Dist" shows the work distance in mm.

Figure 7 is a representative set of photographs/radiographs showing the weld produced in run 21W. Figure 7a is a photograph of the front bead, Figure 7b is a radiograph from the start to the center of the weld and Figure 7c is a radiograph from the center to the finish of the weld. Figure 8 is a radiograph showing a tranverse section through the weld joint of run 21W. The weld bead of run 21W is fairly consistent, with the region near the finish exhibiting some irregularity.

C - Heat Treatment

A number of NVEB-welded samples were subjected to a heat treatment in which the samples were heated in a furnace to a temperature in the range from 800-1000 degrees Celsius for a sufficient time to convert the microstructure of the steel samples to austenite. After the conversion to austenite, the samples were permitted to air cool to ambient temperature.

In the as-welded condition, the samples exhibited a columnar microstructure with some porosity/gas holes. After austenitizing and cooling the weld microstructure and the parent metal was comprised of coarse martensite.

Vestiges of columnar structure persisted in the weld, the bottom upset revealed a band of equiaxed ferrite and the top was completely martensitic. There were no structural differences between a 7 and 8 minute austenitization.

Upon analysis, it was found that the welded steel samples exhibited a consistent microstructure throughout, including across the weld joint. Figure 9 shows an electron beam welded sample before and after austenization for 7 minutes, followed by cooling. The welds appeared satisfactory with minor porosity/gas holes, and with a uniform HAZ of about 3/8 inch width on either side of the weld. Tensile testing resulted in breakage in the parent metal, indicating that the weld is stronger than the parent metal.

It is to be noted that the samples were air cooled as opposed to being quenched with water since an air-cooled sample is more likely to reveal inhomogeneities than a water-quenched sample. It is therefore expected that the results to be obtained with a water quench will be at least as good as those obtained by air cooling, with or without a subsequent tempering step.

Although the invention has been described in connection with certain preferred embodiments, it is not limited thereto. Rather, the invention includes all embodiments which may fall within the scope of the following claims.