Joint method for performing defect-free electron beam welding using the slope-out technique

Abstract

The present invention provides a modification to the EBW process, referred to as the slope-out methodology, which results in the formation of a "slope-out section" generally located within that region of the overall weldment that would exist at the end of a typical EBW welding process for joining two components. The slope-out section overlaps the initial weld on the workpieces for a given distance or length along the weld, effectively eliminating keyholes and, specifically, providing a weldment with minimal or no defects within the slope-out section. The slope-out methodology begins by adjusting various parameters associated with the electron beam to essentially attenuate the beam. Generally, the focal position of the electron beam is transitioned from an under-focused condition (where the focal position is within the material bulk) to an over-focused condition (where the focal position is ahead of the workpiece surface) as the overlap weld is formed.

Description

Technical Field

[0001] The present invention relates to electron beam welding and methods for eliminating keyhole-like portions at the end of a welding process, including its various embodiments. Specifically, the present invention relates to a method for adjusting certain parameters used in an electron beam welding process to join two parts or components at the end of welding, close the keyhole-like portion, and provide a completed weld with minimal or no defects, including its various embodiments.

Background Art

[0002] Electron beam welding (EBW) has the potential to dramatically shorten the welding time for joining thick-section components, such as pressure-retaining components, used in equipment that holds fluids under pressure, such as pressure vessels. These components have thick sections or thick walls to provide the strength required to hold fluids under pressure, and thus require welding or joining of these thick-section components. In some embodiments, these thick-section components are circular and require circumferential welding, as in the manufacture of pressure vessels. EBW can significantly shorten the overall time required to join two components while joining materials up to about 8 inches (200 mm) thick in a single weld pass. Therefore, EBW is an attractive welding method for joining such materials.

[0003] However, during welding, the concentrated electron beam used in EBW penetrates the substrate material and forms a keyhole-like portion, which is a hole in the material being welded located at the leading edge of the weld pool. This keyhole-like portion exists throughout the entire welding process, including up to the end point of welding. For linear or circumferential (pipe or shell) welding, the result will be a keyhole-like portion at the end of the weld, which is essentially a hole through the workpiece, and is not acceptable, especially in the case of pressure-retaining components.

[0004] Furthermore, in some cases, EBW results in defects within the weld, particularly defects near keyholes, which will subsequently require repair. In the context of larger pressure-retaining components, typically with thicker walls, the presence of such defects reduces the attractiveness of EBW for welding such components, especially in light of safety concerns when using high-pressure-retaining equipment such as pressure vessels.

[0005] Therefore, there is a need for an EBW (Electrowelding-Based Welding) method that closes the keyhole-shaped portion at the end of the welding process, and in particular produces welds with minimal or no defects within the area where the keyhole-shaped portion is closed. In particular, there is a need for such a welding process for pressure-retaining components such as pressure vessels, which require circumferential welding of thick-walled segmented components. [Overview of the Initiative] [Means for solving the problem]

[0006] In general, the present invention relates to electron beam welding (EBW) and methods for eliminating keyholes at the end of a welding process used to join two parts or components. Specifically, the present invention, including various embodiments thereof, relates to methods for adjusting welding parameters near the end of the welding process to close keyholes, and in particular to providing a weld with minimal or no defects, such as pores including spike pores, within the area where the keyhole is closed. The present invention generally provides a modification to the EBW process, referred to as the "slope-out methodology," which results in the formation of a "slope-out portion" located within that area of ​​the entire weld where it is located at the end of a normal EBW process. The slope-out portion overlaps with the initial weld of the workpiece over a given distance or length along the weld, effectively filling the keyhole and completing the weld, and in particular providing a weld with minimal or no defects within the slope-out portion. The slope-out methodology can be used in linear or circumferential welding using EBW and is particularly useful in welding thick-walled components.

[0007] The slope-out portion of a weld is generally generated by using a slope-out methodology, which is initiated near the end of the overall welding process or near the end of the initial weld being completed. Prior to the initiation of the slope-out methodology, the workpiece is welded in a steady-state manner with the electron beam focal position or focal plane located within the bulk of the material being welded (i.e., the beam is under-focused or negatively defocused) and the electron beam parameters are kept constant. The slope-out methodology is initiated by adjusting various parameters related to the electron beam to attenuate the beam and form a weld that overlaps the initial weld. In some embodiments, the slope-out methodology includes a step of modifying the beam focal position so that it progresses or changes with the slope-out distance or as the slope-out methodology progresses, which can be either a linear or curved relationship and can persist throughout the slope-out methodology. Generally, the electron beam focal position moves from an underfocused or negatively defocused state (where the focal position is within the material's bulk due to steady welding) to an overfocused or positively defocused state (where the focal position is outside the outer surface of the workpiece) as the overlapping weld is formed. In addition, in some embodiments, the electron beam oscillation pattern may also be modified at the start of the slope-out methodology or as the electron beam focal position changes from underfocused to overfocused (or from negative to positive). In some embodiments, the oscillation pattern is broadened. The slope-out methodology continues for a predetermined amount of time or until an overlapping weld of a predetermined length is formed, at which point the keyhole is closed and both the entire welding process and the slope-out methodology are completed.

[0008] The present invention provides a method for closing a keyhole-shaped portion formed by EBW and producing a weld without resulting defects within the slope-out region. This allows the invention to be used not only for joining two components or parts in a single weld pass, but also for joining thick-walled components such as cylindrical components of thick-walled sections. Thus, the absence of defects reduces or eliminates any rework, which would otherwise negatively impact the gain produced by welding in a single pass. The present invention provides, for example, the following: (Item 1) A method for welding two components together, The welding of two components using electron beam welding, which includes an electron beam, is performed over a first period of time. The focus of the electron beam is adjusted continuously and throughout the entire second period from within the height of the two components to above the two components. Includes, A method wherein the second period begins within the first period, and the first and second periods end in parallel. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 shows a reactor pressure vessel welded using EBW (Electromagnetic Wave Welding).

[0010] [Figure 2A] Figure 2A illustrates a workpiece and electron beam welding machine in a 2G welding orientation according to one embodiment of the present invention.

[0011] [Figure 2B] Figure 2B illustrates the keyhole-shaped portion in the EBW.

[0012] [Figure 3] Figure 3 is a flowchart illustrating a slope-out methodology as part of the overall EBW process according to one embodiment of the present invention.

[0013] [Figure 4] Figure 4 illustrates the effect of changes in electron beam defocusing on the keyhole shape.

[0014] [Figure 5] Figure 5 illustrates the results of focusing trials produced with a linearly progressive lens focus position during the slope-out methodology.

[0015] [Figure 6] Figure 6 illustrates a photograph comparing the appearance of welds produced during slope-out for the various defocusing trials of Figure 5.

[0016] [Figure 7] Figure 7 illustrates a photograph of a longitudinal slice of the weld of Figure 6.

[0017] [Figure 8A] Figure 8A illustrates a phased array ultrasonic testing (PAUT) graph for the slope-out weld of Figure 6.

[0018] [Figure 8B] Figure 8B illustrates a phased array ultrasonic testing (PAUT) graph for a portion of the graph of Figure 8A.

[0019] [Figure 9] Figure 9 illustrates a photograph comparing the appearance of welds produced during the degassing test.

[0020] [Figure 10] Figure 10 illustrates a phased array ultrasonic testing (PAUT) graph for the slope-out weld of Figure 9.

[0021] [Figure 11] Figure 11 illustrates a photograph of a longitudinal slice of the weld of Figure 9.

[0022] [Figure 12] Figure 12 illustrates a photograph of the slope-out region of a full circumferential weld.

[0023] [Figure 13]Figure 13 shows the phased array ultrasonic (PAUT) graph of the slope-out weld in Figure 12, and the longitudinal slice of the weld in Figure 12.

[0024] [Figure 14A] Figures 14A and 14B illustrate the results from additional experiments conducted using rings fabricated from SA508 Grade 3. [Figure 14B] Figures 14A and 14B illustrate the results from additional experiments conducted using rings fabricated from SA508 Grade 3.

[0025] [Figure 15] Figure 15 illustrates the results from the second set of bead-on-plate trials.

[0026] [Figure 16] Figures 16A and 16B show photographs of the ring welding setup for welding two 150 mm high SA508 Grade 3, Class 1 forged rings (Test Ring Weld 1).

[0027] [Figure 17] Figures 17A and 17B show photographs of the completed slope area of ​​the weld from the actual weld of two 150 mm high SA508 Grade 3, Class 1 forged rings (test ring weld 1).

[0028] [Figure 18] Figures 18A and 18B show the ToFD inspection results from the slope-out region of test weld 1.

[0029] [Figure 19] Figures 19A and 19B illustrate the PAUT scan for test weld 1.

[0030] [Figure 20] Figure 20 shows the weld resulting from the step of joining two 150 mm high sections of an SA508 Grade 3, Class 1 ring (test ring weld 2).

[0031] [Figure 21] Figure 21 shows the cosmetic welding path applied to test ring welding 2.

[0032] [Figure 22] Figures 22A and 22B illustrate the setup for shell / flange welding.

[0033] [Figure 23] Figures 23A and 23B show photographs of the completed weld and enlarged views of the slope-out area for the shell / flange weld.

[0034] [Figure 24] Figure 24 shows the machining traces for shell / flange welding.

[0035] [Figure 25] Figure 25 shows a photograph of DPI symptoms for shell / flange welding.

[0036] [Figure 26] Figure 26 shows ToFD data from shell / flange welds.

[0037] [Figure 27] Figure 27 shows the PAUT results from shell / flange welding.

[0038] [Figure 28] Figure 28 shows a phased array graph of shell / flange welding. [Modes for carrying out the invention]

[0039] Detailed description of the invention The present invention will be fully described below with reference to the accompanying drawings or figures. The present invention will be described in conjunction with specific embodiments, but these should be considered examples only and not to be considered to limit or describe only embodiments of the present invention. Rather, the present invention includes various embodiments or forms and various related aspects or features and uses, as well as substitutes, modifications and equivalents, which are all included in the spirit and scope of the invention and claims, whether or not they are expressly described herein. Furthermore, the use of the terms “invention,” “present invention,” “embodiments,” and similar terms throughout this description is used broadly and is not intended to mean that the present invention requires or is limited to any particular embodiment or aspect described, or that such description is the only way in which the present invention may be made or used.

[0040] In general, the present invention relates to electron beam welding (EBW) and methods for eliminating keyholes at the end of a welding process used to join two parts or components. Specifically, the present invention, including various embodiments thereof, relates to methods for adjusting welding parameters at the near end of the welding process to close keyholes, and in particular to providing a weld with minimal or no defects or embedded defects, such as pores including spike pores, within the area where the keyholes are closed. The present invention generally provides a modification to the EBW process, referred to as a “slope-out” methodology or process, which results in the formation of a “slope-out portion” located within that area of ​​the entire weld that is located at the end of a normal EBW process. In particular, the slope-out portion is essentially a portion of the weld produced at the end of the welding process that includes the location of the keyholes, overlapping with the weld formed at the disclosure of the welding process. Thus, the slope-out process occurs at some point near the end of the entire welding process, which is the point at which the weld begins to overlap with the weld formed at the beginning of the entire welding process. Subsequently, both the slope-out process and the entire welding process can be considered to be completed in parallel.

[0041] For example, in circumferential welding, the workpiece is rotated 360 degrees during the welding process. Once the workpiece has been fully rotated 360 degrees and welded around its entire circumference, the slope-out process begins, and a slope-out portion is formed. In this case, the slope-out portion would be the part of the weld that overlaps with the initial portion of the weld produced at the disclosure of the welding process when the workpiece first began to rotate. Thus, this slope-out portion extends over a given length or distance along the weld depending on various parameters, as will be further described below. This slope-out portion effectively fills the keyhole-shaped portion, completing the weld, and in particular, providing a weld with minimal or no defects within the slope-out portion. Once the slope-out portion is formed as desired, the slope-out process ends, which occurs in parallel with the end of the entire welding process.

[0042] The slope-out portion of a weld is generally generated by using a slope-out methodology or process, which is initiated near the end of the overall welding process. In other words, the slope-out methodology is performed near the end of the overall welding process, specifically when the weld begins to overlap the initial weld, and continues until the slope-out portion is complete. Because the slope-out process is performed at the end of a normal EBW process, it can be considered the final part of the overall welding process.

[0043] For example, prior to the initiation of the slope-out process, the workpiece is welded in a steady-state manner with the electron beam's focal position or focal plane within the material being welded (i.e., the beam is under-focused or negatively defocused) and the electron beam parameters kept constant. It should be understood that the precise focal position will be based on material and geometric considerations and will form part of the steady-state welding procedure necessary to perform the welding. Generally, referring to circumferential welding, the slope-out process begins when the workpiece has achieved a full 360-degree rotation so that the steady-state weld is completed around almost the entire circumference of the workpiece. The slope-out process then begins when the formation of the initial 360-degree weld is nearly complete or complete. In other words, the slope-out process begins when the beam overlaps the weld created at the start of the welding process (i.e., the start of the workpiece rotation).

[0044] The slope-out process is initiated by adjusting various parameters related to the electron beam, which essentially attenuate the beam and form an overlapping weld to the initial weld, while continuously rotating the workpiece. The slope-out process continues for a predetermined amount of time, or until an overlapping weld of a predetermined length is formed, at which point the keyhole is closed, and both the overall welding process and the slope-out process are completed. In some embodiments, the slope-out portion may extend along the weld for a length of 2 to 12 inches, depending on the thickness of the workpiece and the time required to generate a defect-free slope.

[0045] In some embodiments, the slope-out process includes a step of modifying the lens focal position so that it progresses or changes with the slope-out distance or as the slope-out process progresses. The change in lens focal position can be either linear or curved and can be persistent throughout the slope-out methodology or procedure. In other words, the rate of change in lens focal position or the magnitude of defocusing can change linearly at a constant rate or nonlinearly at various rates as the slope-out methodology or slope-out welding progresses. Generally, during the slope-out process or as the overlapping weld or slope-out portion is fabricated, the electron beam focal position moves from an under-focused or negative defocused state (where the focal position is within the material's bulk due to steady welding) to an over-focused or positive defocused state (where the focal position is ahead of the workpiece surface). Generally, an under-focused beam refers to a focal point within the workpiece, while an over-focused beam refers to a focal point outside the workpiece surface. A sharp focus refers to a focal point on the workpiece surface.

[0046] In addition, in some embodiments, the electron beam oscillation pattern may also be modified at the start of the slope-out process or as the electron beam focal position changes from an under-focused to an over-focused state (or from negative to positive). In some embodiments, the oscillation pattern is broadened. In some embodiments, an elliptical oscillation pattern, stretched perpendicular to the welding direction, is used. In this case, the elliptical pattern is further stretched at the start of the slope-out process. In some embodiments, the elliptical oscillation pattern is stretched both vertically and in the welding direction. Such broader oscillations increase the size of the keyhole cavity and reduce or avoid the formation of characteristic defects such as spike defects. However, it should be understood that such changes or modifications to the oscillation should be avoided in order to destabilize the keyhole. In general, it should also be understood that the scale of defocusing and the size of the oscillation also depend on the beam geometry.

[0047] Generally, the successful slope-out condition can be achieved depending on the geometry of the workpiece, including, for example, the thickness and radius / length of the material and the rate of change of the focal position; however, the final focal position is the distance from the material surface. The successful slope-out condition is also a function of the electron gun geometry, beam geometry, beam current, beam acceleration voltage, and operating distance, and can be derived from them.

[0048] The slope-out procedure is achieved by balancing three key factors or beam characteristics that occur simultaneously during the slope-out process. First, the beam current is continuously reduced during the slope-out region, which effectively reduces the resulting beam current in welding, as the process transitions from full-penetration welding to partial-penetration welding. Second, the beam focal position is manipulated from a position within the material's bulk for steady-state welding (under-focused) to a position above the workpiece surface (over-focused). An over-focused beam is required to control partial-penetration welding. Third, the beam oscillation is manipulated from horizontal oscillation (parallel to the beam direction) to vertical oscillation (perpendicular to the beam direction). This oscillation counteracts the sharpening of the electron beam profile as a result of the beam power reduction occurring during slope-out. This oscillation is also used to avoid "spike" type defects.

[0049] In all cases, the response of the material or workpiece to changes in beam characteristics has a certain inherent inertia. Therefore, the rate of change of these critical parameters (beam current, focal position, and beam oscillation) must occur over a sufficiently long period to ensure that defects are not "frozen" in the material as a result of rapid solidification. The length of time depends on the welding conditions (e.g., length and rate of welding) as well as the material properties (e.g., solidification range and boiling point of the main element). In some embodiments, allowing a slope-out of about 10 times the thickness of the weld provides sufficient time for the melt pool flow to adapt to the change. In some embodiments, allowing a slope-out of one-tenth the thickness of the weld may be sufficient. For example, if large pores are trapped, this would indicate that the period is too short. Alternatively, if large pores are not trapped, it may be possible to shorten the period. Thus, performing a slope-out process and allowing the material to equilibrium will provide a better weld. Ensuring that the material has time to equilibrium during the slope-out process means either that the length of the slope-out portion or area is sufficiently long (e.g., fixed welding speed), or that the welding speed is reduced during the slope-out process.

[0050] Therefore, the time required for the slope-out process, which can be determined by either the length of the weld or the welding speed adjustment, can be determined iteratively, for example, by using the NDT technique as a guide for weld quality. However, it should be understood that longer lengths or slower welding speeds are advantageous for defect-free slope-outs. It should also be understood that as the thickness of the workpiece increases, the time required for the slope-out process should similarly be extended. This can be achieved by extending the length of the slope-out portion or area, or by reducing the welding speed between the slope-out portion or area.

[0051] It should be understood that many different workpieces can be welded using the EBW and methods of the present invention described herein, and that the invention is particularly well suited for circumferential welding. It should be understood that the invention can be used in welding various materials with varying thicknesses, referred to as thick-walled section welding. It should also be understood that the methods of the present invention can be used when welding workpieces having various construction materials. For example, in some embodiments, the workpiece may be carbon steel, low-alloy steel, nickel steel, or stainless steel. In some embodiments, the construction material or composition of the workpiece will determine the steady-state electron beam welding conditions, and the slope-out process conditions will then be determined as described above, starting from the steady-state welding conditions. However, the physical properties of the workpiece can be used to aid in determining the slope-out process conditions. For example, materials such as nickel alloys result in a more viscous molten pool and flow that is virtually less favorable. In such embodiments, the slope-out process can be slowed down, for example, by forming longer slope-out portions or regions. The solidification range of the workpiece composition is another factor that may be used. For example, materials that solidify more rapidly will have a greater tendency to trap pores. In such embodiments, a slope-out process may be implemented to form longer slope-out portions or regions. However, materials with relatively faster solidification will be beneficial for bead stability.

[0052] Furthermore, it should be understood that the method of the present invention can be implemented without the need for additional equipment or components other than those typically used in EBW. In other words, the method of the present invention can be used in conjunction with existing electron beam welding machines, and their various components, including focusing lenses or magnetic lenses, and refractive systems, whether they are magnetic or implemented by moving the workpiece, but without the use of additional equipment or components added to the electron beam welding machine. In other words, the method of the present invention can be implemented by adjusting welding parameters, including defocusing, without the need for additional components or equipment to otherwise modify the operation of the electron beam welding machine.

[0053] In the following, accompanying figures are discussed in addition to further details of the method of the present invention. The following description provides information on identifying parameters that are useful in implementing the method of the present invention to achieve defect-free electron beam slope-out. In addition, results from various experimental tests are provided to demonstrate the method of the present invention and illustrate the identification of parameters that are useful in implementing the method of the present invention.

[0054] Figure 1 shows a reactor pressure vessel welded using EBW according to one embodiment of the present invention. The reactor pressure vessel 102 is welded to a flange 104. The weld 106 is a circumferential weld produced using EBW. This illustrates one embodiment of a workpiece welded using EBW. It should be understood that many different workpieces can be welded using the EBW and method of the present invention described herein, and that the invention is particularly well suited for circumferential welding. It should also be understood that the use of the EBW and method of the present invention is particularly suitable for welding thick materials, such as materials having a thickness of about 25 to 200 mm (1 to 8 inches) or more, in some embodiments. In some embodiments, the method of the present invention is particularly suitable for welding components used in equipment having thick-walled partitioned components that operate under pressure or retain pressure.

[0055] Figure 2A illustrates a workpiece and electron beam welder in a 2G welding orientation according to one embodiment of the present invention. As shown, the workpiece 202, which may be a reactor pressure vessel having thick-walled segmented components to be welded, is oriented vertically to enable welding. The electron beam welder 204 is oriented horizontally and includes an electron beam head 206 and a rotor 208. This orientation of the workpiece 202 and electron beam welder 204 is a 2G orientation and is useful in circumferential welding, i.e., where the workpiece is rotated 360 degrees during steady-state welding. In this orientation, it should be understood that the workpiece is rotated and the electron beam welder remains in a steady state during welding. Other orientations or positions of the workpiece relative to the electron beam welder can also be used in conjunction with the method of the present invention, so it should be understood that the use of EBW using the method of the present invention is not necessarily limited to the use of the 2G welding position.

[0056] Figure 2B illustrates a keyhole-shaped portion formed in the EBW process. As shown, the EBW process is used to join the workpiece 214. The EBW process forms a plasma 216, which results in the formation of a melting chamber 218. The EBW process causes the formation of a keyhole-shaped portion 212 within the joined workpiece 214. The keyhole-shaped portion 212 is an opening in the workpiece 214 that is closed or filled by the slope-out methodology of the present invention.

[0057] Figure 3 is a flowchart illustrating a slope-out methodology as part of an overall EBW process according to one embodiment of the present invention. The slope-out methodology can be used for either circumferential or linear welding, but the following will be described in terms of circumferential welding. As described above, the slope-out methodology is started near the end of the overall welding process. Thus, the EBW process will be carried out first (302). The EBW process will be carried out under steady-state conditions, as determined by the material being welded and the EBW equipment. To achieve circumferential welding, the electron beam may be introduced into the material or workpiece with increased beam power, which results in the initial formation of a keyhole shape. While the workpiece is rotated a full 360 degrees, the beam welding parameters are then kept essentially constant, and welding continues around the entire circumference of the workpiece. It should be understood that in some cases the workpiece is rotated and the electron gun (and therefore the electron beam) is in a steady state, or vice versa. Furthermore, it should be understood that the precise focal position of the electron beam will be based on material and geometric considerations, and will form part of the steady-state welding procedure necessary to perform the welding. In some embodiments, the electron beam focal position is in or under-focused within the material being welded.

[0058] Once fully rotated, the slope-out methodology will be initiated (304). As described above, the slope-out methodology begins when the weld is formed around the entire circumference of the workpiece. In other words, the slope-out methodology begins when the electron beam begins to overlap the weld formed at the start of the steady-state welding process.

[0059] The slope-out methodology begins by adjusting various parameters related to the electron beam to continuously rotate the workpiece while attenuating the beam, thereby forming an overlapping weld to the initial weld (304). Generally, the adjustments may be made to the magnetic focusing lens amplitude and the beam oscillation amplitude. In some embodiments, the slope-out methodology includes a step (306) of modifying the lens focal position so that it changes in either a linear or nonlinear manner as the slope-out methodology progresses. Generally, the electron beam focal position moves from an under-focused or negative defocused state (where the focal position is within the material's bulk due to steady welding) to an over-focused or positive defocused state (where the focal position is ahead of the workpiece surface, or where the focal position is at a higher level within the workpiece) as the overlapping weld is formed. In addition, the electron beam oscillation pattern may also be modified (308). In some embodiments, the oscillation pattern is broadened. In some embodiments, an elliptical oscillation pattern, stretched perpendicular to the welding direction, is used. In some embodiments, an elliptical oscillation pattern is used, stretched perpendicular to the welding direction. In some embodiments, the elliptical oscillation pattern is stretched perpendicular to the welding direction and linearly in the welding direction. In this case, the elliptical pattern is further stretched at the start of the slope-out process. Such wider oscillations increase the size of the keyhole cavity and reduce or avoid the formation of characteristic defects such as spike defects.

[0060] Figure 4 illustrates the effect of changing the electron beam defocusing on the keyhole shape. The x-axis (horizontal) represents the defocusing distance (mm), ranging from -100 mm on the left to 100 mm on the right. The y-axis (vertical) represents the distance between the electron beam and the workpiece surface. Each shape from left to right illustrates the keyhole shape over a beam diameter of 1 mm and a Rayleigh length of 20 mm. The horizontal arrows for each keyhole shape indicate the electron beam focal position. For each shape, the lighter shaded area on the left indicates the direction of the electrons, and the darker shaded area on the right indicates the intensity contour. Figure 4 shows how the electron beam shape changes as the focal position moves from an under-focused state at the beginning of the slope-out region (left side of Figure 4) to an over-focused state at the end of the slope-out region (right side of Figure 4).

[0061] Figure 4 clearly shows that, to demonstrate the most efficient use of beam energy, further penetration (reaching 100 mm in this case) is achieved through the use of an under-focus adjustment position. This is desired, for example, for the initial weld around the entire circumference in circumferential welding, and the ideal focal position depends on the beam diameter and Rayleigh length as well as the thickness and properties of the material. Figure 4 also shows that a reduced diameter area is generated within the weld, which should be avoided during the slope-out methodology. As the slope-out methodology is initiated, the beam is defocused so that the focal region moves to different levels within the thickness of the workpiece, higher relative to the workpiece or closer to the electron beam gun. In other words, the beam needs to be defocused to an over-focus adjustment state along the entire length of the slope-out. As a result, gas and porosity accompanying porosity at its thickness level along with the workpiece can be reduced or eliminated.

[0062] Returning to Figure 3, the slope-out methodology continues for a predetermined time as the workpiece continues to rotate, or for a given distance to produce an overlapping weld that is a given distance along the initial weld, starting from the beginning of the workpiece's rotation. Again, the slope-out methodology is implemented at approximately the end of the entire welding process, at the point when the slope-out portion of the entire weld begins to overlap with the weld produced immediately before during the same entire welding process, and continues until the slope-out portion is formed. For example, the entire welding process may take a given period of time, and the slope-out methodology may take a shorter period within the longer overall period for the entire welding process. The period of the slope-out methodology will therefore overlap with the same period at the end of the entire welding process, so that both periods may end in parallel.

[0063] During this period for the slope-out methodology, the defocusing of the electron beam lens continues based on the desired rate and magnitude of defocusing to be achieved. Again, generally, defocusing progresses from an under-focused state to an over-focused state during the slope-out methodology. It should be understood that the length of the overlap (slope-out region) is determined by the thickness of the welded components. Thicker sections may have longer slope-outs, while thinner sections may have shorter slope-outs.

[0064] Once the slope-out methodology has formed an overlapping weld over a desired distance along the initial weld or over a desired time while the workpiece is rotated, both the slope-out methodology and the entire welding process will be completed (308). Thus, at this point, the weld is completed without defects and the keyhole is closed.

[0065] Generally, the successful slope-out condition is achieved according to the geometry of the part (material thickness and radius / length) based on the rate of change of the focal position, and the final focal position is located at a distance from the material surface. In addition, the successful slope-out condition is also a function of the gun geometry, beam geometry, beam current, beam acceleration voltage, and working distance. Therefore, different materials will weld differently and require different welding parameters. However, the slope-out methodology of the present invention can be used with any material welded using EBW. In some embodiments, it should be understood that materials with higher thermal conductivity will have less penetration for the same power / velocity. Therefore, a focal position that is more under-focused may not be required for one material than for another. Furthermore, the rate of over-focusing adjustment for its high thermal conductivity material during the slope-out methodology will need to be higher than that for its lower thermal conductivity material.

[0066] It should be understood that the present invention can be used in welding various materials, including SA508 Grade 3, Class 1 and 2 steel that has undergone both forging and hot isostatic pressing (HIP) treatment, which can be used, for example, in thick-walled compartmentalized pressure vessels. [Examples]

[0067] The following are the results of several experiments conducted to evaluate the effects of the slope-out methodology and various parameters related to the slope-out portion of the weld. A 1,960 mm (OD) × 80 mm (wall thickness) S355 (low-carbon manganese steel) ring was used (see Figure 1 above). The welding parameters were developed using an elliptical electron beam oscillation (2 × 1 mm) that was elongated relative to the welding direction, which produced a more stable weld than the same oscillation pattern that was elongated in the welding direction. A 100 mm length full penetration weld was followed by a 400 mm length slope-out region. The welding parameters are presented in Table 1. [Table 1]

[0068] For all trials, the steady-state welding current (dependent on the start of the slope-out methodology) was 450 mA, and the welding defocusing current was -490 mA, which effectively placed the electron beam focus at a depth of 20 mm (from the outer diameter) within the material's bulk. During the slope-out methodology, the welding current was linearly reduced from the steady-state welding condition (450 mA) to 0 over the length of the slope-out (400 mm).

[0069] Figure 5 illustrates the results of focusing trials produced as the lens focal position progresses linearly during the slope-out methodology. In Figure 5, the slope-out progresses from right to left (backward from 450 mA to 0 mA), and the various rates at which defocusing occurs are shown by individual lines, which gradually range from 490 (lowest line) to 1,130 (highest line). The change in defocusing from 450 mA to 0 mA has the effect of moving the electron beam focal position from an under-focused state (the focal position is within the material's bulk due to steady welding) to a forward state. Therefore, for low lens current trials (490 mA), the electron beam focal position is essentially on the material surface at the end of the slope-out, while for higher lens current trials, the beam becomes over-focused (the focal point is in front of the material surface). The higher the lens current, the earlier the beam focal point moves from inside the material to a certain distance away from the material surface. In some embodiments, the decompression position can be +2 to +4 times the initial steady-state weld decompression position, depending on the material, gun geometry, and power.

[0070] Figure 6 illustrates photographs comparing the appearance of welds produced during slope-outs for various defocusing trials in Figure 5, performed using a bead-on-plate method. Similarly, the weld progresses from right to left, following a 100 mm long steady-state weld shown on the right (indicated by the "Weld" arrow at the top), where the slope-out methodology begins (to the left of the "Weld" arrow, at the location indicated by the "Slope-out" arrow at the top). The vertical arrows in each individual figure of the weld identify the point where the electron beam was focused on the surface of the workpiece. The portion of the slope-out weld where the beam was under-focused is illustrated by the horizontal "UF" range (indicated by the UF arrow), and the portion of the slope-out weld where the beam was over-focused is illustrated by the horizontal "OF" range (indicated by the OF arrow).

[0071] All instances preceding (or to the right of) the slope-out arrow were under-focused, and all instances following it were over-focused (i.e., the beam's maximum power density was outside the material, i.e., towards the gun). Under the 1,130 mA condition, the weld began to exhibit a drop at the top of the weld, and therefore no further trials were performed. Low lens current trials (490 mA) are associated with the general case where the focal position returns to 0 during the slope-out. The fact that the weld becomes tapered at the end of the slope indicates that the beam is effectively focused on the surface at the end of the slope (indicated by the arrow). As can be seen, the weld molten pool is unstable and drips at the end of the slope.

[0072] During the slope-out, increasing the defocusing value at +570, +650, +730, +810, and +970 mA significantly improved the appearance of the slope-out region. At +570 mA, the molten pool was unstable, and a drop at the top was observed. In this situation, the focus was again located near the edge of the slope. However, as the defocusing value increased, the point where the beam focus was precisely on the top surface of the material appeared earlier than during the slope-out (indicated by the arrow). This is visible through the narrowing of the weld top. At the 1,130 mA condition, the weld began to exhibit a drop at the top, and therefore, no further trials were performed. This indicates that an initial defocusing rate of +1,130 was better, but this was too high and led to a decline, illustrating that in some embodiments, a nonlinear defocusing rate of the slope-out methodology, i.e., bilinear or higher, may be preferable.

[0073] Manual ultrasonic testing (UT) was performed to evaluate defects. Using this process, the parameter set with a slope focus of (-490) + 970 mA was the condition that yielded the fewest defects. In addition, this slope-out condition produced welds with the best appearance. Based on these results, this parameter set was selected for the subsequent investigations described below.

[0074] Figure 7 illustrates a photograph of a longitudinal slice of the weld in Figure 6. Because the slope-out weld overlapped perpendicularly with the original or initial weld within the slope-out area, it was necessary to inspect the upper weld and then pulverize and remove the material to reach the next weld below it. As shown in Figure 7, it was possible to reach the weld area and visually observe any defects through continuous machining.

[0075] Figure 8A illustrates the phased array ultrasonic (PAUT) graph for the slope-out welds in Figure 6. Note that the horizontal marks extending across most of the horizontal portion of the graph are noise. The sloped portions represent the slope-out areas. Figure 8B illustrates the phased array ultrasonic (PAUT) graph for a portion of the graph in Figure 8A. For the three welds in Figure 8 (welds performed at 490, 650, and 970 mA), Figure 8B identifies the start of the slope-out portion (identified by the "Slope-Out Start" arrow) and various defects (identified by arrows associated with the specifically described defects). Table 2 records the defect information as a function of the welding input parameters. [Table 2]

[0076] Two defect types were observed. Generally, pores were observed that were set at the beginning of the slope-out (see the circle at 650 in Figure 6). Also observed following the weld fusion line were spike defects, which are at the interface where the slope-out weld is above the original weld and can be seen as an upward-sloping "line" progressing from right to left in Figure 6. The spike defects are represented as a sharp increase in electron beam transmission beyond the mean transmission line.

[0077] As highlighted in Figures 8A and 8B and Table 2, spike defects appeared to decrease as the defocusing value increased. As a result, spike defects were almost completely eliminated by the 1,130 mA condition. This appears to indicate that beams in overfocused configurations do not suffer as many spike defects as those in underfocused configurations. Referring specifically to Figure 8B, as the focal position shifts from 490 to 970, the overall results shift to smaller spike events. In addition, the frequency and severity of the spikes decrease as the focal position shifts.

[0078] Based on the previous results, several welds were performed using strategies that increased the effective degassing time. Degassing during welding helps avoid porosity and shrinkage-type defects, and also helps suppress spike defects. These strategies included increasing the slope length (effectively slowing down the slope-out methodology) by using a wider oscillation parameter, and increasing the keyhole cavity size. Reducing the welding speed not only reduced the time for degassing but also resulted in defective welds, and therefore this approach was not pursued.

[0079] Figure 9 illustrates photographs comparing the appearance of multiple welds produced during the degassing test. Figure 10 illustrates a phased array ultrasonic (PAUT) graph for the slope-out weld in Figure 9. Table 3 provides a list of welding parameters examined. [Table 3]

[0080] Weld 1 was a repeat of the 970mA weld from the test described above, but with a longer slope length (800mm vs. 400mm above). It was evident that spike defects were still present but significantly reduced (see Figure 10, Weld 1). Weld 2 included an additional modification in which the beam oscillation was increased to 3 × 1 mm. As can be seen from Figure 10, the phased array ultrasonic inspection results for welds 1 and 2 appeared to be similar to each other.

[0081] The increase in beam oscillation to 3 mm across 800 mm was deemed too slow. Therefore, for weld 3, the beam oscillation was increased to 4 × 1 mm. As can be seen from Figure 10, the phased array ultrasonic inspection results for weld 3, there was one small pore defect remaining at the beginning of the slope-out region, otherwise the slope-out would have been completely defect-free. This was similar for weld 4, where the oscillation expansion was applied to both dimensions of the beam (4 × 2 mm oscillation).

[0082] Figure 11 illustrates photographs of longitudinal slices of the weld in Figure 9. These highlight the machined surface of the slope-out. Several defects are visible in the photographs, as indicated by circles within welds 1 and 3. Weld 1 shows a spike defect, and weld 3 shows a pore (approximately 3 mm in diameter). Despite continuous cutting in 0.2 mm increments, the spike defects were so minute on the scale that they were difficult to observe.

[0083] Figure 12 illustrates a photograph of the slope-out region of a full-circumference weld. Figure 13 illustrates a phased array ultrasonic (PAUT) graph of the slope-out weld in Figure 12 and a longitudinal slice of the weld in Figure 12. The weld shown in Figure 12 is an equivalent of weld 3 described above and was produced by performing a full-diameter weld, which lasted for 70 minutes around the entire diameter of an 80 mm wall-thick ring with an OD of 1,960 mm, followed by a slope of 800 mm in length. The phased array ultrasonic graph in Figure 13 shows that the slope-out is defect-free. The longitudinal slice in Figure 13 is marked to show how the electron beam was extracted or defocused during the slope-out methodology.

[0084] S355 steel behaved differently from SA508 in terms of electron beam welding characteristics. With respect to the same parameters, the S355 material resulted in a much flatter apex and a more protruding root. This can be observed by the amount of relief groove on the weld surface. This relief groove was also visible at the beginning of the more protruding overlap. Except for the two relief groove points shown in Figures 12 and 13, the electron beam weld slope-out contained no defects.

[0085] Additional experiments were conducted using rings fabricated from SA508 Grade 3, a low-alloy steel. A first series of bead-on-plate welds was performed over a total length of approximately 500 mm, consisting of a 100 mm steady-state weld followed by approximately 400 mm of slope-out welds. The welding parameters are as shown in Table 4. [Table 4]

[0086] During slope-out welding, the welding current was linearly reduced from steady-state welding conditions (450 mA) to 0 mA over the length of the slope-out (400 mm for these trials). These trials aimed to assess the effect of changing the electron beam welding focal point as the current was reduced. Five slope-out trial welds were performed to assess the effect of reducing the welding power from the initial maximum steady-state conditions to 0 mA at the end of the slope-out.

[0087] The primary parameter assessed during the welding trials was the welding defocusing parameter, as shown in Table 5. Increasing the defocusing parameter (from +490 to +1,290 mA) effectively means that the beam focal length is moving further away from the workpiece surface over the same slope-out length. In other words, the focus is moving from the material towards the EB gun at a faster rate, as described above. [Table 5]

[0088] Figures 14A and 14B illustrate the results from additional experiments performed using rings fabricated from SA508 Grade 3. These figures include the appearance of five trial welds for volume assessment and the corresponding PAUT scans. The vertical lines on each photograph of the weld indicate the narrowing of the weld bead, specifying the sharp focal position corresponding to the beam focus on the surface of the workpiece material. Boxes are superimposed on each PAUT scan to indicate the weld. Each column is identified as AE by the amount of defocusing.

[0089] The appearance of the weld during the slope-out was a critical indicator of weld quality. A defocusing parameter of +490 mA resulted in a general loss in the apex region immediately after the point where the steady-state weld transitioned to a slope-out (see Figure 14A, column A). Weld apex loss was also observed over the slope-out, characterized by maximum defocusing (+1,290 mA), which occurred closer to the end of the slope-out region. Both the +730 mA and +970 mA defocusing conditions produced very stable welds (in terms of appearance). The sharp focal positions, indicated by separate vertical lines in each photograph, suggest that the focal progression is relatively stable for SA508 grade 3 substrate material with respect to +730 mA and +970 mA.

[0090] PAUT scanning further confirmed this at +490mA, +1,130mA, and +1,290mA, showing pore / cavity-type symptoms that should be rejected. However, the +730mA condition showed improvement but still included spike defects of interest from the PAUT scan. The +970mA condition yielded the best PAUT result, with a small area corresponding to the spike-type defects of interest. Table 5 also highlights these results.

[0091] A second set of bead-on-plate trials used the promising +970mA condition (from above) to assess the increase in slope-out distance, and also included the application of beam oscillation pattern deflection (from horizontal to vertical). Details of these trials are recorded in Table 6. [Table 6]

[0092] Figure 15 illustrates the results from a second set of bead-on-plate trials. This figure includes the appearance of the weld and the PAUT inspection results. Boxes are superimposed on each PAUT scan to indicate the weld. The first slope-out weld (A) was virtually a repeat of the initial weld and presented a very similar appearance. The second slope-out weld (B) maintained all other process parameters while virtually increasing the slope length from 400 to 800 mm. PAUT scans from this weld showed a reduction in the severity of spike defects, indicating that increasing the slope-out length was beneficial. Note that if this weld were assessed against ASME V criteria, the spike defects would be well below the reportable symptom size.

[0093] The third slope-out trial (C) demonstrates a transition in beam oscillation from a fully transmitted oscillation pattern (horizontal) to a partially transmitted oscillation pattern (vertical). The slope-out length was maintained under the same conditions as described above (slope-out length = 800 mm). PAUT scanning reveals additional improvements in the quality of the slope-out region. Again, under ASME V standards, this weld would be considered defect-free.

[0094] The fourth slope-out trial (D) maintained the oscillation change and further increased the slope-out length from 800 to 1,600 mm. The results from this weld were excellent, and the PAUT results showed that the slope-out was clean, perfect, and free of any defects.

[0095] It should be understood that the results suggest that slope-out length is indeed a very important factor in preventing pore / defect trapping. Furthermore, the application of oscillation modification (horizontal to vertical) reduced instances of spikes.

[0096] A complete welding slope-out procedure was developed for welding a 2 / 3 scale reactor pressure vessel protrusion. Table 7 presents the welding parameters. This slope-out procedure was applied to an actual thick-walled section weld joint produced from SA508 Grade 3. [Table 7]

[0097] The first weld (test ring weld 1) was performed by joining two 150 mm high SA508 Grade 3, Class 1 forged rings together. Both mating rings were separated from the same forging. Figures 16A and 16B show photographs of the ring weld setup for welding the two 150 mm high SA508 Grade 3, Class 1 forged rings (test ring weld 1), which includes 5 mm start and stop pre- and post-bars (not shown) to control weld drip. Figures 17A and 17B show photographs of the completed slope area of ​​the weld from the main weld (test ring weld 1) of the two 150 mm high SA508 Grade 3, Class 1 forged rings.

[0098] Figures 18A and 18B show the ToFD inspection results from the slope-out region of test weld 1. The results suggested the presence of a series of symptoms representing electron beam transmission as it traced the slope-out line and moved effectively outside and away from the material. These symptoms began at the back wall (0 mm position and inner diameter) and continued to the outer diameter. Note that the results indicated coverage only from the 0 mm (back wall) position to 42 mm from the outer surface. Figure 18B shows the marked section within the box. It is important to note at this stage that the symptoms were not measured via ToFD.

[0099] ToFD scanning was assisted by PAUT scanning across the same slope-out region. Figures 19A and 19B illustrate the PAUT scan for test weld 1. Boxes 1902, 1904, 1906, and 1908 indicate the weld. Figure 19A shows the collected, unfiltered dates, and Figure 19B shows the filtered dates in accordance with AME detection requirements. The collected raw data (Figure 19A), employing high amplitude signal gain, suggests the presence of several symptoms that effectively trace the slope-out line. However, the application of the ASME V inspection criteria (Figure 19B) demonstrates that the symptoms are well below reportable size. This suggests that the slope-out region was acceptable and no reportable symptoms were present. It is noteworthy that no reportables were observed in the remainder of the weld during the inspection (initial and steady state).

[0100] A second weld (test ring weld 2) was performed as a direct repeat of weld 1 by joining two 150 mm high sections of an SA508 Grade 3 Class 1 ring (again from the same forging). The goal was to determine whether the weld slope-out was reproducible using another set of forgings. Similar procedures were used for the slope in test ring weld 2 and in the slope-out. Figure 20 shows the weld resulting from the step of joining two 150 mm high sections of an SA508 Grade 3, Class 1 ring (test ring weld 2), and is also visually similar. Both ToFD and PAUT presented nearly identical results for test ring weld 2 compared to test ring weld 1. Figure 21 shows the decorative weld pass applied to test ring weld 2. The decorative pass was successful in reducing the occurrence of insufficient filling of the weld face.

[0101] A third weld was performed on a full-scale model of the lower section of the NuScale reactor pressure vessel lower assembly, consisting of a forged flange, a shell cylinder, and a recessed lower head. The shell / flange weld consists of a forged flange (outer diameter 1,780 mm) welded to a forged shell. The joint thickness is 80 mm and includes an integrated 5 mm step on the inner and outer diameters to act as a weld support (similar to a pre-bar). The flange was centered using DTI, and the flatness of the joint was also corrected. Figures 22A and 22B illustrate the setup for the shell / flange weld.

[0102] The welding program was semi-automated using seam tracking and automated tacking procedures. This activity took 15 minutes. Welding was then completed in 50 minutes (beam-on time). Figures 23A and 23B show photographs of the completed weld and enlarged views of the slope-out area for the shell / flange weld, respectively. The entire operation, from setup to completion of welding, took 2 hours.

[0103] Figure 24 shows the machining trace for shell / flange welding, which records the voltage and current during electron beam welding. Some high-voltage instability up to 3kV is shown, resulting in some arc generation errors. However, it should be emphasized that the welding proceeded as planned.

[0104] The ID and OD of the shell / flange were machined to remove the weld apex and root, as well as the post-bar, from the front and rear locations. This was required to facilitate complete NDT of the weld. For machining, up to 5 mm was removed from both sides to clean the weld. After machining, dye penetration testing (DPI) was performed.

[0105] Figure 25 shows a photograph of DPI symptoms for shell / flange welding. While the DPI showed no symptoms in the ID, the OD presented four isolated areas corresponding to underfill. These areas required subsequent manual dressing to a depth of 4 mm to clean them up. The most pronounced symptom corresponded to the slope within the area with underfill at the beginning of the slope-in / slope-out overlap. There is no evidence in these areas to be related to high-voltage / beam-current events, such as those recorded from the k2000 trace. These features would benefit from using a cosmetic pass, as described above.

[0106] Figure 26 shows ToFD data from a shell / flange weld, and Figure 27 shows PAUT results from a shell / flange weld. The same volumetric NDT approach, combining both ToFD and PAUT, was employed for the shell / flange weld as described above. For the shell / flange weld, only Figures 26 and 27 show scanning from the slope-out region. The rest of the weld did not present any recordable symptoms. The ToFD data from the shell / flange weld shows that the symptoms in the slope region were the same as those observed above in welds 1 and 2. Note that the scanning was obtained from the inner diameter, and therefore the symptoms were in the opposite direction to the results described above. Examining the PAUT results, the scanning revealed no recordable symptoms during the slope-out portion of the shell / flange. Figure 28 shows the phased array graph of the shell / flange weld. Once the ASME V standard was applied to the scanning, the trace became very clean, as shown at the bottom of Figure 28. These results confirm that the slope-out procedure is transferable because the shell / flange welds are from different temperatures of SA508 Grade 3 material.

[0107] Various embodiments of the present invention have been described above. However, it should be understood that alternative embodiments are possible and that the present invention is not limited to the specific embodiments described above.

Claims

1. A method for welding two components together, The welding of two components using electron beam welding, which includes an electron beam, is performed over a first period of time. The adjustment of the focus of the electron beam, continuously and throughout the entire second period, from within the bulk of the two components to a focal position above the two components and closer to the electron beam, wherein the adjustment of the focus of the electron beam includes a nonlinear rate of change. During the second period, the shape of the oscillation pattern of the electron beam is adjusted. Includes, A method wherein the second period begins within the first period, and the first and second periods end in parallel.

2. A method for welding two components together, The welding of two components using electron beam welding, which includes an electron beam, is performed over a first period of time. The focus of the electron beam is adjusted continuously and throughout the entire second period from within the height of the two components to a focal position above the two components and closer to the electron beam. During the second period, the shape of the oscillation pattern of the electron beam is adjusted, wherein the oscillation pattern comprises an elliptical pattern, and adjusting the shape of the oscillation pattern includes stretching the elliptical pattern. Includes, A method wherein the second period begins within the first period, and the first and second periods end in parallel.

3. The method according to claim 2, wherein the elliptical pattern is stretched perpendicular to the direction of welding.

4. The method according to claim 3, further comprising extending the elliptical pattern in the direction of welding.

5. A method for circumferential welding of two components using electron beam welding, The method involves forming a first welded portion using an electron beam, wherein the electron beam extends from a first position along the joint between two components to a second position adjacent to the first position. A slope-out weld is formed by forming a second weld that overlaps with the first weld formed at the first position by a predetermined length, The focus of the electron beam is continuously adjusted, while the slope-out weld portion is being formed, from within the height of the two components to a focal position above the two components and closer to the electron beam. During the formation of the slope-out weld portion, the shape of the electron beam oscillation pattern is adjusted, wherein the oscillation pattern comprises an elliptical pattern, and adjusting the shape of the oscillation pattern includes stretching the elliptical pattern. Methods that include...

6. The method according to claim 5, wherein the elliptical pattern is stretched perpendicular to the direction of welding.

7. The method according to claim 6, further comprising extending the elliptical pattern in the direction of welding.

8. A method for welding two components together, The process of welding a joint between two components using an electron beam, wherein the welding involves forming a weld having an initial portion formed when the welding is initiated. To form a slope-out weld portion that overlaps with the initial portion of the welded portion, The focus of the electron beam is continuously adjusted, while the slope-out weld portion is being formed, from within the height of the two components to a focal position above the two components and closer to the electron beam. During the formation of the slope-out weld portion, the shape of the electron beam oscillation pattern is adjusted, wherein the oscillation pattern comprises an elliptical pattern, and adjusting the shape of the oscillation pattern includes stretching the elliptical pattern. While forming the slope-out weld portion, the electron beam current is continuously reduced. Methods that include...

9. The method according to claim 8, wherein the elliptical pattern is stretched perpendicular to the direction of welding.

10. The method according to claim 9, further comprising extending the elliptical pattern in the direction of welding.

11. The method according to claim 1, further comprising closing the keyhole-shaped portion before the end of the second period.

12. The method according to claim 2, further comprising closing the keyhole-shaped portion by the end of the second period.

13. The method according to claim 5, further comprising closing the keyhole-shaped portion before the completion of forming the slope-out weld portion.

14. The method according to claim 8, further comprising closing the keyhole-shaped portion before the completion of forming the slope-out weld portion.

Images