Laser welding of metallic pin pairs using time-dependent scanning patterns and energy inputs

Abstract

A method for laser welding a pair of metal pins (182, 200) includes delivering a laser beam (112) to a work side (220) of the pair of metal pins, where the respective pair of surfaces of the metal pins are adjacent to each other and face the same direction. The laser beam first traces a first path (230, 630) on the work side, forming a molten pool (850) by keyhole welding. The first path intersects the interface (210) between the metal pins. After tracing the first path, the laser beam is switched to trace a second path (240, 640) on the work side using a laser beam with a delivery rate of energy per unit path length less than that used for the first path. The second path intersects the interface and is within the first path. This method is well suited to welding hairpins and I-pin stators.

Description

[Technical field]

[0001] (Technical field of the invention) The present invention relates generally to laser welding of metal pin pairs, and more specifically, hairpins and I-pins for rectangular wire stators. [Background technology]

[0002] (Discussion of Background Art) Laser welding uses a continuous or pulsed laser beam as a concentrated heat source to locally melt and join two parts, typically made of metal. The laser beam can be focused to a relatively small spot, resulting in a high power density and a small heat-affected zone. Laser welding is therefore an attractive technique when precision and a high degree of control are required. Furthermore, laser welding lends itself well to automation.

[0003] In laser welding, a focused laser beam precisely positions each weld point or line while minimizing incidental heating. It is useful to distinguish between two main laser welding techniques. Conduction welding occurs at lower laser powers and lower laser power densities. The absorbed laser power heats the irradiated material, causing the material of each part to be joined to melt, which flows, mixes, and then solidifies. Keyhole welding is performed at higher laser powers and higher laser power densities that are sufficient to vaporize a portion of the irradiated material. The pressure that the vaporized material exerts on the surrounding molten material opens a channel through the molten material. This channel, known in the art as a keyhole, has a characteristic narrow and deep profile that allows for deep penetration of the laser beam. The completed keyhole weld is generally narrower, deeper, and stronger than a conduction weld.

[0004] Laser beam welding has been used to weld hairpin stators. Generally, a "stator" has a series of electrical windings that exchange energy with a magnetic rotor. Together, the stator and rotor form the basis of electric motors and generators. In an electric motor, current is passed through the windings of the stator to generate a rotating magnetic field that drives the magnetic rotor. In a generator, the stator converts the rotating magnetic field of the magnetic rotor into an electric current. Hairpin stators use thick rectangular bar wire for their windings. The rectangular bar wire can be packed into the stator with a higher packing factor than conventional wound round wire. Stators with rectangular bar wire are therefore the preferred solution for motors that require high torque and power density in a relatively compact package. In addition, rectangular bar wire is very suitable for liquid cooling. However, the thick rectangular bar wire cannot be wound continuously like conventional round wire. Instead, many hairpin-shaped sections of square wire rod are inserted into slots in the stator ring from one side, with the ends of the hairpins extending out the other side of the stator ring. Adjacent ends are then welded together to form a continuous winding from the individual sections. In some cases, so-called I-pins are used in place of hairpins. Essentially, each hairpin is replaced by two I-pins, and welding of the exposed ends is required on both sides of the stator ring. Summary of the Invention [Means for solving the problem]

[0005] SUMMARY OF THEINVENTION Disclosed herein is a method for laser welding of metal pin pairs. The method is particularly well suited for welding hairpins or I-pins of stators, since the method (a) minimizes or even eliminates spatter during the welding process, (b) prevents or reduces the formation of voids in the weld nugget, and (c) maximizes the depth and area of ​​the weld nugget along the interface between the two metal pins. Spatter is an undesirable effect in keyhole welding, where metal evaporation is intense enough that metal droplets are ejected during the welding process. The ejection of these droplets reduces the volume of the weld nugget in an irregular and poorly controlled manner. Spatter therefore adversely affects the electrical conductivity through the weld, thus at least partially negating the advantage of using square wire rods for stators. The formation of voids is another undesirable effect in keyhole welding, where bubbles of metal gas cannot escape to the surface before the molten metal solidifies. The voids impede the electrical conductivity of the weld nugget. The ability to pass electrical current from one of the metal pins to another is also affected by geometric aspects of the weld nugget. For example, the larger the cross-sectional area of ​​the weld nugget at the interface between the two metal pins, the better the overall electrical conductivity between the two metal pins. The method applies a time-dependent laser scanning pattern and energy input to address these issues.

[0006] In one aspect, a method of laser welding a pair of metal pins includes the steps of: (a) delivering a laser beam to a work side of the pair of metal pins, where surfaces of each pair of metal pins are adjacent to each other and face in the same direction; (b) tracing a first path on the work side with the laser beam at a first delivery rate of energy per unit path length to form a molten pool by keyhole welding, where the first path crosses the interface between the metal pins; and (c) after tracing the first path, tracing a second path on the work side with the laser beam at a second delivery rate of energy per unit path length that is up to 2 / 3 of the first delivery rate of energy per unit path length, where the second path crosses the interface and is within the first path. [Brief description of the drawings]

[0007] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate preferred embodiments of the invention and, together with the general description above and the detailed description of the preferred embodiments below, serve to explain the principles of the invention.

[0008] [Figure 1] FIG. 1 illustrates one laser welding apparatus for welding metal pin pairs according to one embodiment.

[0009] [Diagram 2] 2A-C show a pair of metal pins and the elliptical path traced by a laser beam during laser welding of the pair of metal pins, according to one embodiment.

[0010] [Diagram 3] FIG. 3 is a flow chart for a method for welding a pair of metal pins, according to one embodiment.

[0011] [Figure 4] FIG. 4 shows an example of a metal pin pair laser welded according to the method of FIG.

[0012] [Diagram 5] FIG. 5 shows an example of a metal pin pair that has been laser welded according to a conventional laser welding process.

[0013] [Figure 6] FIG. 6 shows an example of a closed path that replaces the elliptical path of FIG. 2B.

[0014] [Figure 7] FIG. 7 shows an embodiment of a hybrid closed-open path that replaces the elliptical path of FIG. 2B.

[0015] [Figure 8] 8A-E are a series of diagrams illustrating one embodiment of laser welding of a metal pin pair according to the method of FIG.

[0016] [Figure 9] 9A and 9B show the growth of the molten pool during one step of the method of FIG.

[0017] [Figure 10] FIG. 10 is a diagram of the keyhole aspect ratio of depth versus width.

[0018] [Figure 11] FIG. 11 shows a pair of metal pins joined together with their respective end faces facing in opposite directions, according to one embodiment.

[0019] [Figure 12] FIG. 12 shows one embodiment of a metal pin pair joined to a non-flat work side for laser welding according to the method of FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] Detailed Description of the Invention Referring now to the drawings, where like components are designated with like numerals, FIG. 1 shows one laser welding apparatus 100 for welding metal pin pairs. In an example of a usage scenario depicted in FIG. 1, the laser welding apparatus 100 is welding pairs of metal pins 182(1) and 182(2) of a stator 180. Each metal pin 182 may be a hairpin or an I-pin. A more detailed view of some of the hairpin pairs 182(1, 2) is provided in an enlarged view 190 of a portion of the stator 180. The laser welding apparatus 100 includes a laser source 110 and a fixture 120. The laser source 110 generates a laser beam 112. In one embodiment, the laser beam 112 is either pulsed or continuous wave, and the laser source 110 is capable of delivering an average power of at least 1 kilowatt. The laser beam 112 may be near infrared or visible (e.g., green). The laser welding apparatus 100 is configured to move the laser beam 112 relative to the stator 180 to sequentially weld each pair of metal pins 182(1,2), with the laser beam 112 tracing a time-dependent scan pattern on the working side of each pair of metal pins 182(1,2). In one embodiment, the fixture 120 rotates the stator 180 in stages to place the metal pin pairs 182(1,2) one by one at the focal point of the laser beam 120, and the laser source 110 uses a beam scanner to trace a time-dependent scan pattern on the working side of each metal pin pair 182(1,2). In another embodiment, all the movements are performed by the laser source 110. For example, the location of the beam scanner of the laser source 110 may be shifted and directed to each metal pin pair 182(1,2) in sequence, with the beam scanner then being manipulated to trace a time-dependent scan pattern on the working side of each metal pin pair 182(1,2).

[0021] 2A-C show a pair of metal pins 200(1) and 200(2). Metal pin pair 200(1,2) is a hairpin or I-pin of a stator, such as metal pin pair 182(1,2) of stator 180. Most commonly, stators use copper pins. Thus, metal pins 200(1) and 200(2) may be made of copper. Alternatively, metal pins 200(1) and 200(2) may be made of aluminum or another metal. FIG. 2A shows metal pin pair 200(1,2) in a perspective view. FIG. 2B is a top plan view of working side 220 formed by end faces of metal pins 200(1) and 200(2). FIG. 2C is a side cross-sectional view of metal pin pair 200(1,2) taken as indicated by line 2C-2C in FIG. 2A. A Cartesian coordinate system 290 illustrates the orientation of the metal pin pair 182 (1, 2) in each of Figures 2A-2C. The working side 220 faces in the positive z-axis direction. Each metal pin 200 may extend in the negative z-axis direction beyond what is shown in Figures 2A and 2C, such that Figures 2A and 2C show only an end portion of each metal pin 200. Additionally, outside the views of Figures 2A-C, the non-end portion of each metal pin 200 may be bent, for example, as is customary for hairpins and I-pins.

[0022] Metal pins 200(1) and 200(2) are positioned adjacent to one another and form an interface 210 therebetween. Interface 210 is parallel to the yz plane. Metal pins 200(1) and 200(2) may directly contact one another at interface 210. A gap between metal pins 200(1) and 200(2) may exist in all or part of interface 210 without departing from the scope of this specification. As shown in FIG. 2B, working side 220 is formed by end faces 204(1) and 204(2) of metal pins 200(1) and 200(2), respectively. FIGS. 2A-C show each end face 204 as being planar and end faces 204(1) and 204(2) as being coplanar with one another. For example, hairpins and I-pins are typically cut from long rods of square wire, and the cutting process may leave ridges on the end of each pin and produce other non-planarity. Additionally, the positioning of hairpins and I-pins may be subject to manufacturing tolerances.

[0023] Each metal pin 200 has a width 202W along the x-axis and a length 202L along the y-axis. In one embodiment, the width 202W is in the range of 1 to 3 millimeters (mm) and the length 202L is in the range of 3 to 10 mm. Each end face 204 may have a rounded rectangular shape, a true rectangular shape, or a non-rectangular trapezoidal shape, as shown in Figures 2A-C.

[0024] 3 is a flow chart for one method 300 for welding a pair of metal pins. Method 300 may be used to weld metal pin pair 200 (1, 2) and may be performed by laser welding apparatus 100. Method 300 moves a laser beam over the working side of the metal pin pair in a time-dependent scan pattern and at a time-dependent energy delivery rate. The time-dependent scan pattern and energy delivery rate help minimize or eliminate spatter and maximize the area of ​​the weld interface.

[0025] 4 and 5 show metal pin pair 200 (1, 2) welded according to method 300 and a conventional laser welding method, respectively. FIG. 4 shows metal pin pair 200 (1, 2) laser welded according to method 300 in the same cross-sectional view as used in FIG. 2C. Method 300 forms a weld nugget 470 that intersects interface 210 and has a depth 474 along interface 210 (along the z-axis). FIG. 5 shows metal pin pair 200 (1, 2) laser welded according to a conventional laser welding process in the same cross-sectional view as used in FIG. 2C. The conventional laser welding process forms a weld nugget 570 that intersects interface 210 and has a depth 574 along interface 210.

[0026] The weld nugget 570 has some voids 572 caused by metal bubbles entrapped in the molten metal during the conventional laser welding process. In the case of the weld nugget 470, the method 300 eliminates or at least minimizes the presence of voids. By optimizing melting at the interface 210 and reducing or eliminating the loss of weld nugget material due to spatter, the method 300 further maximizes the depth 474 of the weld nugget 470 such that the depth 474 of the weld nugget 470 exceeds the depth 574 of the weld nugget 570. The voids 572 of the weld nugget 570 are not electrically conductive and therefore impede the flow of electrical current 590 between the conventional welded metal pins 200(1) and 200(2). The shallower depth 574 of the weld nugget 570 may further impede the flow of electrical current 590. In contrast, the current 490 through the weld nugget 470 is not impeded (or at least only minimally impeded) by the voids, and the depth 474 may be sufficient so as not to impede the current 490 beyond the impedance associated with the cross-sectional area of ​​each metal pin 200 perpendicular to the flow of the current 490.

[0027] 3, the method 300 includes steps 310, 320, and 330. Step 310 delivers a laser beam to a working side of the metal pin pair, where individual pairs of surfaces of the metal pin pair are adjacent to one another and face in the same direction. In one example of step 310, the laser source 110 delivers the laser beam 112 to the working side 220 of the metal pin pair 200(1,2).

[0028] Step 320 traces a first path on the working side to form a molten pool by keyhole welding. The first path intersects the interface between the metal pins. In step 320, the laser beam power and the speed at which the laser beam moves along the first path correspond to a first delivery rate of energy per unit length of the first path. Here, the delivery rate of energy refers to the energy of the laser radiation incident on the metal part, regardless of whether the laser radiation is absorbed or reflected. In one embodiment of step 320, the laser beam 112 traces a path 230 (shown in FIG. 2B) on the working side 220 of the metal pin pair 200 (1, 2). The laser beam 112 may make one, several, or partial revolutions on the working side 220 along the path 230. After performing step 320 for a duration, the method 300 switches to step 330.

[0029] Step 330 traces a second path on the working side. The second path crosses the interface between the metal pins and is within the first path. Here, the term "within the first path" encompasses scenarios where the first path does not complete a full 360 degree revolution on the working side 220. In such scenarios, the second path is within an extrapolation of the first path, a full revolution of 360 degrees. In step 330, the laser beam power and the speed at which the laser beam moves along the second path correspond to a second delivery rate of energy per unit length of the second path. The second delivery rate of energy is lower than the first delivery rate of energy of step 320. In one embodiment of step 330, the laser source 110 traces a path 240 (shown in FIG. 2B) on the working surface 220 of the metal pin pair 200 (1, 2). The laser beam 112 typically makes several revolutions on the working side 220 along the path 240. Path 230 is more peripheral than path 240, and path 240 is within path 230. The shape of paths 230 and 240 may be different than the ellipse shown in FIG. 2B. Each of paths 230 and 240 may be a closed path (e.g., an ellipse, a circle, or a rectangle) or an open path (e.g., a spiral). When the path is an ellipse, the major axis of the elliptical path may be aligned with the interface between the metal pins or aligned with the longer dimension of the working side.

[0030] Figure 6 shows one alternative closed path to the elliptical path of Figure 2B, where a first path 630 is a rounded rectangle with its major axis aligned along the interface 210, and a second path 640 is a circle centered at the interface 210.

[0031] Figure 7 shows one hybrid closed-open path alternative to the elliptical path of Figure 2B, where the elliptical second path 240 is replaced with a spiral second path 740 that spirals inward from the first path 230.

[0032] Many other scanning patterns are possible. For example, the first and second paths may be (a) successive portions of the same inward spiral, (b) concentric circles, or (c) concentric or rounded rectangles.

[0033] 8A-E are a series of diagrams illustrating an example of laser welding of a metal pin pair 200 (1, 2) according to method 300. This example uses elliptical paths 230 and 240, but is applicable to other shapes of first and second paths, such as those discussed above with reference to FIGS. 6 and 7. Each of FIGS. 8A-E is a top plan view of the working side 220 of the metal pin pair 200 (1, 2) at a separate stage of method 300. In step 320, the laser beam 112 traces the path 230 (see FIG. 8A). Keyhole welding by the laser beam 112 along the path 230 results in the formation of a molten pool 850, which initially tracks the laser beam 112 as it moves along the path 230. During step 320, the melt pool 850 grows (see FIG. 8B) and after a while it no longer resembles the local tail-tracking laser beam 112, but rather forms a common melt pool that covers a substantial portion of the area within and defined by the path 230 (see FIG. 8C).

[0034] Switching from step 320 to step 330 involves two changes: (a) decreasing the rate of delivery of energy per unit path length, and (b) moving the energy delivery by the laser beam 112 to a more central region of the working side 220 (see FIG. 8D). The decrease in the rate of delivery of energy per unit path length helps, at least in part, to avoid or reduce spatter to prevent or minimize loss of material from the metal pin pair 200 (1, 2). During step 330, the laser beam 112 continues to melt the metal pin pair 200 (1, 2) to grow the molten pool 850 (see FIG. 8E). The rate of delivery of energy per unit path length in step 330 is sufficient for keyhole welding, however, at a level that avoids or minimizes spatter. The reduction in the rate of delivery of energy per unit path length when switching from step 320 to step 330 may be achieved by increasing the speed at which the laser beam moves across the work surface and / or by reducing the power of the laser beam. Thus, step 330 may include one or both of: (i) step 334, tracing a second path at a faster speed than that used for the first path in step 320; and (ii) step 336, using a lower laser beam power than in step 320. For practical simplicity and a more easily controlled process, an increase in speed may be preferred over a reduction in laser beam power. Thus, in one embodiment, method 330 includes step 334 and omits step 336, such that the laser beam power is identical in steps 320 and 330.

[0035] The movement of the laser beam 112 to the more central area serves to preferentially grow the cross-sectional area of ​​the molten pool 850 at the interface 210 while benefiting, at least for a time, from the confinement of the molten pool 850 by the more unmelted peripheral portions of the metal pin pair 200 (1, 2).

[0036] 9A and 9B illustrate the growth of the melt pool 850 during step 330. FIGS. 9A and 9B are cross-sectional views of a metal pin pair 200(1,2) similar to that used in FIG. 2C, respectively. As shown in FIG. 9A, while the laser beam 112 travels along the path 240, thus heating the melt pool 850 from a relatively central location, the unmelted peripheral portions 956 of the metal pins 200(1) and 200(2) form a barrier that helps prevent the melt pool 850 from spilling over the edge of the working side 220, at least for a time. Eventually, the growth of the melt pool 850 may include the peripheral portions 956, and the melt pool may grow to a final size covering the entire working side 220, as shown in FIG. 9B. The relatively central location of the laser beam 112 in step 330 maximizes the depth 374 of the final melt pool 850, and therefore the weld nugget 470 (and maximizes the general cross-sectional area of ​​the weld nugget 470 along the interface 210). If the laser beam 112 instead proceeds along the path 230 and the rate of delivery of energy per unit path length decreases, the molten material near the periphery of the work side 220 may be lost and splashed or spilled around, so that the melt pool 850 does not achieve the same final depth 474 along the interface 210.

[0037] In application of method 300 applied to a sample pair of metal pins made from copper, it has been found that best results are obtained when the lateral dimension of the second path along the working side is between 20% and 80% of the corresponding lateral dimension of the first path along the working side. Referring to FIG. 2B, in one embodiment of method 300, path 230 has a width 232W in the x dimension and a length 232L in the y dimension, and path 240 has a corresponding width 242W and length 242L, where (a) length 232L is between 60% and 90% of the length 202L of each metal pin 200, width 232W is between 50% and 90% of twice the value of width 202W of each metal pin 200, and (b) each of length 242L and width 242W is between 20% and 80% of length 232L and width 232W, respectively.

[0038] Referring again to FIG. 3, the reduction in the delivery rate of energy per unit path length between steps 320 and 330 may further help reduce or eliminate void formation by limiting the keyhole depth. Metal bubbles are more likely to become trapped when they form away from the surface. Thus, a shallower keyhole reduces the risk of void formation. A shallower keyhole also reduces the risk of sputtering. Thus, steps 320 and 330 may include steps 322 and 332, respectively, that maintain the keyhole aspect ratio of depth to width below a threshold value.

[0039] FIG. 10 is a diagram illustrating the depth-to-width keyhole aspect ratio. Laser beam 112 is incident on working side 220 on metal pins 200(1) and 200(2) and travels along working side 220 as shown by arrow 1090. Laser beam 112 forms keyhole 1060 in melt pool 850, with the majority of the volume of melt pool 850 following keyhole 1060. Keyhole 1060 has a depth 1062D in the negative z-axis direction from the surface of working side 220. The depth-to-width keyhole aspect ratio is depth 1062D divided by width 1062W. Width 1062W may vary along the z-axis, but width 1062W generally correlates to the beam size of laser beam 112 incident on working side 220. Therefore, for most practical purposes, the depth-to-width keyhole aspect ratio is approximately the depth 1062D divided by the diameter of the laser beam 112 (e.g., the waist diameter of the laser beam 112 when focused on the working side 220).

[0040] It has been found that depth-to-width keyhole aspect ratios greater than 3.0 are prone to spatter and / or void formation. This discovery was based on an investigation of several metal pin pairs 200 (1, 2) made from copper using a combination of real-time imaging of spatter behavior, real-time optical coherence tomography to assess keyhole depth, and post-weld cross-sections. Thus, in one embodiment, the threshold value for steps 322 and 332 is 3.0. In another embodiment to further reduce the risk of spatter, the threshold value for steps 322 and 332 is less than 3.0. In this embodiment, the threshold value may be 2.5, 2.0, or 1.5. Depending on this threshold value and the tolerance for spatter, the depth-to-width keyhole aspect ratio is maintained at a value less than 3.0, less than 2.5, less than 2.0, or less than 1.5 throughout the method 300.

[0041] Steps 322 and 332 have the same purpose but are associated with different requirements, since the working side and molten pool conditions are generally different between steps 320 and 330. In one embodiment, step 322 includes maintaining a first rate of delivery of energy per unit length at or below a first maximum value, and step 332 includes maintaining a second rate of delivery of energy per unit length at or below a second maximum value that is only a percentage of the first maximum value for the first rate of delivery of energy per unit length in step 322. It has been found that under certain circumstances in an embodiment of method 300 including at least steps 322 and 332 and applied to a pair of metal pins made of copper, the second maximum value in step 332 is about 2 / 3 of the first maximum value in step 322. In a corresponding embodiment, the second rate of delivery of energy per unit length in step 330 is at most 2 / 3 of the first rate of delivery of energy per unit length in step 320. In one implementation of this embodiment, step 330 performs step 334 but does not perform step 336 to achieve the desired reduction in the rate of delivery of energy per unit length such that the laser beam traces the second path at a speed at least 1.5 times faster than the speed used for the first path in step 320.

[0042] Moreover, it has been found that, at least under certain circumstances, in an embodiment of method 300 including steps 322 and 332 and applied to a pair of metal pins made of copper, it is preferable to set the second rate of delivery of energy per unit length in step 330 to at least 1 / 5 of the first rate of delivery of energy per unit length in step 320. Thus, in one embodiment of method 300, the second rate of delivery of energy per unit length in step 330 is 1 / 5 to 2 / 3 of the first rate of delivery of energy per unit length in step 320. In one implementation of this embodiment, step 330 performs step 334 but does not perform step 336 to achieve the desired reduction in the rate of delivery of energy per unit length. In this implementation, the laser beam traces the second path in step 330 at a speed that is 1.5 to 5 times faster than the speed used for the first path in step 320. The appropriate speed used for the first path in step 320 depends on other parameters, in particular the laser beam power. In one embodiment, the speed used for the first pass in step 320 is in the range of 100-600 millimeters per second and the laser beam power is in the range of 1-10 kilowatts.

[0043] Method 300 may utilize one or more criteria for when to switch from step 320 to step 330. In one embodiment, method 300 implements criterion 390 to switch to step 330 before the depth-to-width keyhole aspect ratio reaches a predetermined threshold. This embodiment of method 300 may include steps 322 and 332, and criterion 390 may be based on the same threshold used in steps 322 and 332.

[0044] As an alternative to or in conjunction with criterion 390, method 300 may implement criterion 392, which switches from step 320 to step 330 before the keyhole reaches a position on the first path that is already occupied by a melt pool. It has been found that if the keyhole (e.g., keyhole 1060) is allowed to travel the first path long enough for the melt pool to grow in size such that the keyhole overtakes the existing portion of the melt pool, the keyhole depth essentially instantly exceeds the limit at which sputtering and / or void formation occurs. Thus, as shown by way of example in FIG. 8C and indicated by intersection 890, criterion 392 imposes a switch to step 330 before laser beam 112 reaches a position on path 230 that is already occupied by a previously formed portion of melt pool 850. In some scenarios, switching from step 320 to step 330 according to criterion 392 also ensures that criterion 390 is met. 3, method 300 may include imaging the working side in real time to identify when it is necessary to switch from step 320 to step 330 according to criteria 392. Method 300 may implement both criteria 390 and 392.

[0045] Criteria 390 and 392 both impose an upper limit on the duration of step 320. Method 300 may further implement criterion 394, which imposes a lower limit on the duration of step 320, by switching to step 330 only after the melt pool has grown to occupy a certain threshold percentage of the area within and defined by the first pass. In one embodiment, the threshold percentage is 50%. In another embodiment, the threshold percentage is greater than 50%, such as in the range of 60%-80%. Although not shown in FIG. 3, method 300 may include imaging the work side in real time to identify when it is permissible to switch from step 320 to step 330 according to criterion 394.

[0046] In the absence of real-time imaging capabilities to assess when to switch from step 320 to step 330 according to one or more of criteria 390, 392, and 394, such tests may be performed on one or more samples to determine machine parameters that ensure switching according to one or more of criteria 390, 392, and 394. For example, such tests may determine appropriate values ​​for one or both of (i) the duration on the first path, and (ii) the rate of delivery of energy per unit length along the first path. These tests may also determine appropriate values ​​of machine parameters required in addition to step 322, such as (i) the size of the first path, (ii) the rate of delivery of energy per unit length along the second path, and (iii) the size of the second path. Thus, method 300 may implement predefined machine parameters to perform and switch between steps 320 and 330 according to one or more of criteria 390, 392, and 394.

[0047] In an embodiment, the duration of step 330 exceeds the duration of step 320 to prevent spatter in step 320 while allowing substantial growth of the molten pool in step 330. The total welding time, including steps 320 and 330, may be less than 200 milliseconds or less than 100 milliseconds.

[0048] The method 300 is not limited to metal pin pairs having the same geometry as the metal pin pair 200 (1, 2). Rather, the method 300 is generally applicable to edge welding of metal parts, and is particularly well suited for welding of rectangular wire rods. Rectangular wire rod pairs or pin pairs may be joined with geometries different from that of the metal pin pair 200 (1, 2), so long as two distinct surfaces of the two metal parts are adjacent to each other and form working sides for welding.

[0049] FIG. 11 illustrates a pair of metal pins 1100(1) and 1100(2) that are joined such that their respective end faces 1104(1) and 1104(2) face in opposite directions. Metal pins 1100(1) and 1100(2) may be ends of respective hairpins or I-pins, as depicted in FIG. Metal pin pair 1100(1,2) represents one example of a pin pair geometry that differs from that of metal pin pair 200(1,2). A working side 1102 is provided by two adjacent side surfaces 1108(1) and 1108(2) of metal pins 1100(1) and 1100(2), respectively, for welding in accordance with method 300. In one embodiment of the method 300 , as applied to the metallic pin pair 1100 ( 1 , 2 ), the laser beam 112 first traces a larger path 1130 in step 320 , and then traces a smaller path 1140 in step 330 .

[0050] 12 illustrates, in a side cross-sectional view similar to that used in FIG. 2C, an example of a metal pin pair joined with a non-planar working side for laser welding according to method 300. In this example, two metal pins 1200(1) and 1200(2) terminate in respective ridges 1206(1) and 1206(2), resulting in a V-shaped working side 1220. The respective portions of working side 1220 formed by metal pins 1200(1) and 1200(2) are still considered to face in the same direction, in the sense that both portions generally face upward in FIG.

[0051] The present invention has been described above in terms of preferred and alternative embodiments. However, the present invention is not limited to the embodiments described and depicted herein. Rather, the present invention is limited only by the claims appended hereto.

Claims

1. 1. A method for laser welding a pair of metal pins, comprising: delivering a laser beam to working sides of the pairs of metal pins, the surfaces of the respective pairs of metal pins being adjacent to one another and facing in the same direction; tracing a first path on the work side with the laser beam at a first delivery rate of energy per unit path length to form a molten pool by keyhole welding; after tracing the first path, tracing a second path on the working side with the laser beam at a second delivery rate of energy per unit path length that is up to ⅔ of the first delivery rate of energy per unit path length, the second path crossing an interface and being within the first path; A method comprising:

2. 2. The method of claim 1 including switching from tracing the first path to tracing the second path before an aspect ratio of a keyhole depth to width of the keyhole weld reaches a value of three.

3. 3. The method of claim 2, further comprising maintaining the depth-to-width aspect ratio less than 3 throughout the steps of tracing the first path and tracing the second path.

4. The method of claim 2 , further comprising maintaining a depth-to-width aspect ratio of less than 2 throughout the steps of tracing the first path and tracing the second path.

5. The method of claim 2 , wherein the molten pool spans across the interface before switching from tracing the first path to tracing the second path.

6. 3. The method of claim 2 including switching from tracing the first path to tracing the second path before the keyhole reaches a position on the first path occupied by the molten pool.

7. The method of any of claims 1 to 6, wherein tracing the second path comprises increasing a depth of the molten pool at the interface.

8. The method according to any one of claims 1 to 7, wherein the pair of metal pins are stator hairpins or I-pins.

9. The method according to any one of claims 1 to 8, wherein the pair of metal pins are made of copper.

10. In the step of tracing the first path, the laser beam traces the first path at a first speed and with a first power; 10. The method of claim 1, wherein in the step of tracing the second path, the laser beam traces the second path at a second speed and with the first power, the second speed being faster than the first speed.

11. 11. The method of claim 10, wherein the second speed is at least 1.5 times the first speed, and the laser beam has the same power in tracing the first path and in tracing the second path.

12. The method of claim 11 , wherein the second rate is at most five times the first rate.

13. In the step of tracing the first path, the laser beam has a first power; The method according to any of the preceding claims, wherein in the step of tracing a second path, the laser beam has a second power lower than the first power.

14. The method of claim 13 , wherein the laser beam traces the first and second paths at the same speed.

15. In the step of tracing the first path, the laser beam has a first power and traces the first path at a first speed; 10. The method of claim 1, wherein in the step of tracing the second path, the laser beam has a second power lower than the first power and traces the second path at a second speed faster than the first speed.

16. The method according to any of the preceding claims, wherein a lateral dimension of the working side of the second pass is between 20% and 80% of a corresponding lateral dimension of the working side of the first pass.

17. The method according to any of claims 1 to 16, wherein a duration of the step of tracing the second path exceeds a duration of the step of tracing the first path.

18. The method according to any one of claims 1 to 17, wherein the first path is a closed path.

19. The method of claim 18 including performing the switching step after the molten pool has grown to occupy at least half of the area within and defined by a closed path.

20. The method of claim 18 , wherein each of the first and second paths is an ellipse.

21. 20. The method of claim 18, wherein the second path is an inward spiral starting from the first path.

22. The method according to any one of claims 1 to 17, wherein the first path is an open path.

Images