Component mounting device

Abstract

The present invention provides a component mounting apparatus comprising: a component recognition unit that performs component position recognition; an inspection unit that inspects whether or not a component can be mounted on a board mounting point; a first control unit that controls operations by the component recognition unit and the inspection unit; and a second control unit that executes a picking setup process to determine the allocation of each of a plurality of components to be picked by which one of a plurality of heads and the picking order of the plurality of components by the plurality of heads. The first control unit causes the component position recognition by the component recognition unit and the inspection of whether or not the component can be mounted by the inspection unit to be performed in parallel. The second control unit executes the picking setup process on the basis of the results of the component position recognition and the inspection as to whether or not mounting is possible which are performed in parallel.

Description

Technical Field

[0001] The present invention relates to a component mounting device that picks up matrix-arranged components with a plurality of heads and mounts them on a substrate.

Background Art

[0002] For example, there is known a component mounting device that picks up matrix-arranged components such as dice of a diced wafer or tray components housed in a tray and mounts them on a substrate (for example, Patent Document 1). In this component mounting device, components carried into the component arrangement area in the machine are imaged by a camera for component recognition, and then each of a plurality of heads provided in a head unit capable of XY movement picks up the components, and the head unit is moved and the components are mounted on the substrate. Before mounting the components on the substrate, an inspection process for determining whether the components can be mounted at predetermined mounting points on the substrate is performed first.

[0003] Conventionally, after performing the above inspection process and discriminating non-mountable mounting points among the mounting points where components cannot be mounted, the component recognition process is executed. Further, after the results of these processes are obtained, a picking setting for determining which of the plurality of heads picks up each of the plurality of components and the picking order is made in one picking operation.

[0004] In the above conventional technology, after inspecting the mounting points, the component recognition and picking setting processes are executed. For this reason, there is a waiting time for the execution of the inspection process, which may deteriorate the tact time.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] The objective of the present invention is to improve the cycle time in a component mounting device that picks matrix-arranged components with multiple heads and mounts them onto a substrate.

[0007] A component mounting apparatus according to one aspect of the present invention comprises: a component supply unit having a component placement area on which a plurality of components arranged in a matrix are arranged; a component recognition unit having a camera for imaging the components in the component placement area and a processing unit that recognizes the position of the components in the component placement area based on the image acquired by the camera; a head unit having a plurality of heads that can move between a component mounting area on which a substrate on which the components are mounted is arranged and the component placement area, and picks the components in the component placement area and mounts the components onto the substrate in the component mounting area; an inspection unit that checks whether components can be mounted on predetermined mounting points on the substrate arranged in the component mounting area; a first control unit that controls the operation of the component recognition unit and the inspection unit; and a second control unit that executes a picking setting process that determines which of the plurality of heads each of the plurality of components is to be picked, and the picking order of the plurality of components by the plurality of heads. The first control unit causes the component recognition unit to perform position recognition of the components and the inspection unit to perform the mounting feasibility check in parallel. The second control unit executes the picking setting process based on the results of the position recognition of the component and the inspection of whether or not it can be mounted, which are performed in parallel. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a top view plan showing the overall configuration of a component mounting device according to an embodiment of the present invention. [Figure 2] Figure 2 is a schematic perspective view showing the head unit and the thrust unit. [Figure 3A] Figure 3A is a top view of a diced wafer. [Figure 3B]Figure 3B shows an example of a wafer map. [Figure 4] Figure 4 is a schematic diagram showing the component mounting status by the head unit on a multi-sided substrate. [Figure 5] Figure 5 is a block diagram showing the control configuration of the component mounting device. [Figure 6] Figure 6 is a flowchart showing the component mounting operation in the comparative example. [Figure 7] Figure 7 is a flowchart showing the component mounting operation according to the embodiment. [Figure 8] Figure 8 is a diagram illustrating the picking setup, including the assignment of the wafer die and suction head, and the mounting points between the die and the substrate. [Figure 9] Figure 9 is a diagram illustrating the change in the picking settings when a mounting point error occurs. [Figure 10] Figure 10 shows an example of changing the picking settings when selecting mount priority mode, which causes the die to be picked by the head adjacent to the head that was assigned to mounting point NG. [Figure 11A] Figure 11A shows the component mounting operation in mount priority mode. [Figure 11B] Figure 11B shows the component mounting operation in mount priority mode. [Figure 11C] Figure 11C shows the component mounting operation in mount priority mode. [Figure 12] Figure 12 shows an example of changing the picking settings when selecting the picking priority mode, which causes the die to be picked by the head that was assigned to mounting point NG. [Figure 13A] Figure 13A shows the component mounting operation in picking priority mode. [Figure 13B] Figure 13B shows the component mounting operation in picking priority mode. [Figure 13C] Figure 13C shows the component mounting operation in picking priority mode. [Figure 14]FIG. 14 is a schematic diagram showing the relationship between the head pitch and the die pitch or the component mounting point pitch, and the amount of movement of the head. [Figure 15] FIG. 15 is a tabular diagram showing the calculation formula for the amount of head movement in each pattern when the picking priority mode is selected. [Figure 16] FIG. 16 is a tabular diagram showing the calculation formula for the amount of head movement in each pattern when the mounting priority mode is selected.

Embodiments for Carrying Out the Invention

[0009] Hereinafter, embodiments of the present invention will be described in detail based on the drawings. The component mounting apparatus according to the present invention is an apparatus that picks up a plurality of components arranged in a matrix, such as dice diced from a wafer, electronic components arranged in alignment in a tray or pallet for accommodating components, by means such as suction, and mounts them on a substrate. Hereinafter, a component mounting apparatus in the case where the component is a die of a wafer will be described.

[0010] [Description of Component Mounting Apparatus] FIG. 1 is a plan view in a top view showing the overall configuration of a component mounting apparatus 1 according to an embodiment of the present invention. The component mounting apparatus 1 mounts a die 7a (component) diced from a wafer 7 on a substrate P. The component mounting apparatus 1 includes a base 2, a conveyor 3, a head unit 4, a component supply unit 5 (component arrangement area), a wafer supply device 6, a camera unit 32U, and a push-up unit 40. FIG. 2 is a schematic perspective view showing the head unit 4 and a push-up unit 40 not shown in FIG. 1.

[0011] The base 2 is the mounting base for various devices provided in the component mounting apparatus 1. The conveyor 3 is a conveyance line for the substrate P installed on the base 2 so as to extend in the X direction. The conveyor 3 conveys the substrate P from outside the machine to a predetermined working position, and after the component mounting operation, conveys the substrate P out of the machine from the working position. The conveyor 3 has a schematic clamp mechanism for holding the substrate P at the working position. In FIG. 1, the position where the substrate P is shown is the working position. In the present embodiment, the working position is a component mounting area where the substrate on which components are to be mounted is arranged. The component supply unit 5 supplies a plurality of dice 7a in a matrix arrangement state diced from the wafer 7.

[0012] The head unit 4 picks up the die 7a in the component supply unit 5, moves to the working position, and mounts the die 7a on the substrate P. The head unit 4 includes a plurality of heads 4H that adsorb and hold the die 7a during the picking, and release the held die 7a during the mounting. The head 4H is capable of moving forward and backward (lifting and lowering) in the Z direction with respect to the head unit 4 and rotating about an axis.

[0013] The head unit 4 is equipped with a substrate recognition camera 31 for imaging the substrate P and a laser height measuring device 4R (a part of the inspection unit). From the captured image of the substrate recognition camera 31, Fid (fiduciary) marks FM1, FM2 and bad marks 66 (FIG. 4) attached to the substrate P are recognized. By recognizing the Fid mark FM, the misalignment of the substrate P is recognized, and the misalignment is corrected during component mounting. Also, by recognizing the bad mark 66, it is possible to inspect whether or not a component can be mounted at a mounting point predetermined as a component mounting position. The laser height measuring device 4R is a device that measures distance using laser light, and is a measuring device for inspecting whether or not there is a foreign object at the mounting point.

[0014] The component mounting device 1 includes a first drive mechanism D1 that enables the head unit 4 to move horizontally (in the X and Y directions) in the space above at least the component supply unit 5 and the substrate P held in the work position. The first drive mechanism D1, as a mechanism for moving the head unit 4 in the Y direction, includes a pair of Y-axis fixed rails 13 on the +X side and the X side, a first Y-axis servo motor 14, and a ball screw shaft 15, respectively. The pair of Y-axis fixed rails 13 are fixed on the base 2 and extend in the Y direction parallel to each other at a predetermined distance in the X direction. The ball screw shaft 15 is positioned to extend in the Y direction at a location close to the Y-axis fixed rails 13. The first Y-axis servo motor 14 rotationally drives the ball screw shaft 15. A support frame 16 supporting the head unit 4 is installed between the pair of Y-axis fixed rails 13. Nuts 17, which are screwed onto each ball screw shaft 15, are assembled to the +X side end and the X side end of the support frame 16.

[0015] The first drive mechanism D1, as a mechanism for moving the head unit 4 in the X direction, includes a guide member (not shown), a first X-axis servo motor 18, and a ball screw shaft 19 mounted on a support frame 16. The guide member is a member that guides the movement of the head unit 4 in the X direction and is fixed to the +Y side of the support frame 16 so as to extend in the X direction. The ball screw shaft 19 is arranged close to the guide member and extends in the X direction. The first X-axis servo motor 18 rotationally drives the ball screw shaft 19. The head unit 4 is fitted with a nut (not shown), which is screwed onto the ball screw shaft 19.

[0016] According to the first drive mechanism D1 having the above configuration, the first Y-axis servo motor 14 is activated and the ball screw shaft 15 is rotationally driven, causing the head unit 4 to move in the Y direction together with the support frame 16. In addition, the first X-axis servo motor 18 is activated and the ball screw shaft 19 is rotationally driven, causing the head unit 4 to move in the X direction relative to the support frame 16.

[0017] The component supply unit 5 includes a wafer supply device 6 that supplies a plurality of dies 7a in the form of a wafer 7 to a predetermined component extraction work position (wafer stage 10). The wafer 7 is a disc-shaped semiconductor wafer on which circuit patterns and the like have already been formed. The wafer supply device 6 includes a wafer holding frame 8 that holds a wafer sheet 8a. The wafer sheet 8a has an assembly of numerous dies 7a, 7a... formed by dicing the wafer 7 in a grid pattern attached to it. The wafer supply device 6 supplies the dies 7a to the component extraction work position by replacing the wafer holding frame 8.

[0018] The wafer supply device 6 includes a wafer storage elevator 9, a wafer stage 10, and a wafer conveyor 11. The wafer storage elevator 9 stores wafer sheets 8a with wafers 7 attached in wafer holding frames 8 in multiple vertical rows. The wafer stage 10 is installed on the base 2 at a position on the -Y side of the wafer storage elevator 9. The wafer stage 10 is positioned on the +Y side with respect to the work position, which is the stopping position of the substrate P. In this embodiment, the wafer stage 108, which is the area on the base 2 where the diced wafers 7 are placed, becomes the component placement area. The wafer conveyor 11 pulls the wafer holding frames 8 from the wafer storage elevator 9 onto the wafer stage 10.

[0019] The camera unit 32U is a unit that can move in the X and Y directions and includes a wafer camera 32 (camera / component recognition unit). The wafer camera 32 images a portion of the wafer 7 positioned on the wafer stage 10, i.e., the component placement area, of the die 7a within the camera's field of view. Based on this captured image, the position of the die 7a to be picked up is recognized. The component mounting device 1 includes a second drive mechanism D2 that enables the camera unit 32U to move horizontally (in the X and Y directions) in the space above at least between the component supply unit 5 and a predetermined standby position. This second drive mechanism D2 is a drive system separate from the first drive mechanism D1 that drives the head unit 4. In this embodiment, the standby position is a position spaced apart from the wafer stage 10 on the +Y side.

[0020] The second drive mechanism D2, as a mechanism for moving the camera unit 32U in the Y direction, comprises a pair of Y-axis fixed rails 33 on the +X side and the X side, and a second Y-axis servo motor 34 and a ball screw shaft 35 located on the +X side. The pair of Y-axis fixed rails 33 are fixed on the base 2 and extend in the Y direction parallel to each other at a predetermined distance in the X direction. The ball screw shaft 35 is positioned to extend in the Y direction at a location close to the Y-axis fixed rail 33 on the +X side. The second Y-axis servo motor 34 rotationally drives the ball screw shaft 35. A support frame 36 supporting the camera unit 32U is installed between the pair of Y-axis fixed rails 33. A nut 37, which is screwed onto the ball screw shaft 35, is assembled to the +X side end of the support frame 36.

[0021] The second drive mechanism D2, as a mechanism for moving the camera unit 32U in the X direction, includes a guide member (not shown), a second X-axis servo motor 38, and a ball screw shaft 39 mounted on a support frame 36. The guide member is a member that guides the movement of the camera unit 32U in the X direction and is fixed to the -Y side of the support frame 36 so as to extend in the X direction. The ball screw shaft 39 is arranged close to the guide member so as to extend in the X direction. The second X-axis servo motor 38 rotationally drives the ball screw shaft 39. The camera unit 32U is fitted with a nut (not shown), which is screwed onto the ball screw shaft 39.

[0022] With the second drive mechanism D2 having the above configuration, the second Y-axis servo motor 34 is activated and the ball screw shaft 35 is rotationally driven, causing the camera unit 32U to move in the Y direction together with the support frame 36. In addition, the second X-axis servo motor 38 is activated and the ball screw shaft 39 is rotationally driven, causing the camera unit 32U to move in the X direction relative to the support frame 36.

[0023] The push-up unit 40 is positioned below the component supply unit 5 and pushes up the die 7a that the head 4H is to pick up from the lower surface of the wafer sheet 8a. The component mounting device 1 of this embodiment is not a flip-flop type, but a direct wafer type device in which the head 4H directly picks up the die 7a that has been pushed up by the push-up unit 40. The push-up unit 40 is positioned on the base 2 so as to be movable in the XY direction over a range corresponding to the wafer stage 10. The push-up unit 40 is supported so as to be movable in the X direction by a support frame 42 which is movable along a pair of guide rails 41 extending in the Y direction.

[0024] A ball screw shaft 43, which screws into a nut portion (not shown) located inside the support frame 42, is rotationally driven by a third Y-axis servo motor 44. This causes the push-up unit 40 to move in the Y direction together with the support frame 42. The support frame 42 also has a ball screw shaft 45, which screws into a nut portion (not shown) located inside the push-up unit 40. The ball screw shaft 45 is rotationally driven by a third X-axis servo motor 46, causing the push-up unit 40 to move in the X-axis direction. The push-up unit 40 has a push-up pin 47 that pushes up the die 7a. When the die 7a is attracted by the head 4H, the push-up pin 47 rises, pushing up the die 7a through the wafer sheet 8a. The push-up pin 47 is raised and lowered by a pin lifting motor 48 (Figure 5).

[0025] A component recognition camera 30 is mounted on the base 2. The component recognition camera 30 captures an image of the die 7a, which is held in place by the head 4H of the head unit 4, from below before it is mounted onto the substrate P. Based on this captured image, any abnormalities or errors in the attachment of the die 7a by the head 4H are determined.

[0026] [Wafer and substrate] Figure 3A is a top view of a diced wafer 7. The wafer 7 contains multiple dies 7a that have been made independent by dicing. The dies 7a are arranged in an X x Y matrix on the wafer sheet 8a. Since the wafer 7 is circular when viewed from above, the number of dies 7a in multiple X rows aligned in the Y direction is higher in the center of the Y direction and lower at the edges. Similarly, the number of dies 7a in multiple Y columns aligned in the X direction is higher in the center of the X direction and lower at the edges. The position of each die 7a is managed by an address based on the XY coordinate system.

[0027] Figure 3B shows an example of a wafer map 7M for wafer 7. The wafer map 7M is a file that describes the evaluation of each die 7a on wafer 7, based on predetermined criteria, whether it is a good or defective product. The evaluation value is described in correspondence to the address of each die 7a. In Figure 3B, "1" indicates a good die 7a, "2" indicates a die 7a that is not defective but of a lower grade, and "3" indicates a defective die 7a. Note that "n" indicates that there is no die 7a at that address. When the head 4H picks up dies 7a from wafer 7, the wafer map 7M is referred to, for example, to control the sequential pickup of only good dies 7a.

[0028] Figure 4 shows an example of a circuit board on which components are mounted by the component mounting device 1. In Figure 4, a multi-chamfered circuit board PA is shown as the circuit board, and the component mounting status on the multi-chamfered circuit board PA by the head unit 4 is schematically shown. The multi-chamfered circuit board PA is a single circuit board made up of a plurality of sub-circuit boards 61 that can be separated from each other, arranged in a matrix. Each sub-circuit board 61 has two mounting areas 63. Multiple predetermined component mounting points are set within the range of these mounting areas 63. Component mounting is performed on a unit of one multi-chamfered circuit board PA, and after component mounting, each sub-circuit board 61 is separated. Perforations or slits are provided between adjacent sub-circuit boards 61 to facilitate separation.

[0029] Figure 4 shows an enlarged view of one sub-board 61. Here, the mounting area 63 is illustrated with IC component mounting points 64 (mounting points) where IC components are mounted, and multiple die mounting points 65 (mounting points) where dies 7a are mounted. Furthermore, bad marks 66 are provided on the periphery of the mounting area 63. If it is determined in the preceding process to the component mounting device 1 that the mounting area 63 is unsuitable for component mounting, the bad marks 66 are, for example, blacked out, as shown in the left-hand mounting area 63. The bad marks 66 are captured by the board recognition camera 31, and the mounting area 63 in which the blacked-out bad marks 66 are recognized is treated as "unsuitable for mounting." In other words, the IC component mounting points 64 and die mounting points 65 of the mounting area 63 are designated as "unsuitable for mounting" and are excluded from component mounting.

[0030] The number of unit mounting areas 62 on the chamfered substrate PA is determined according to the number of heads 4H provided by the head unit 4. Figure 4 illustrates a head unit 4 in which 10 heads 4H are arranged in a straight line. Each head 4H includes a shaft 51 that can move in the vertical direction and a suction nozzle 52 attached to the lower end of the shaft 51. The group of suction nozzles 52 provided by the head unit 4 can pick up 10 parts in one part picking operation at the parts supply unit 5. Assume that each part picked up by each suction nozzle 52 in one picking operation is mounted in one mount area 63. In this case, there are two mount areas 63 for each sub-substrate 61, so the unit mounting area 62 is the area of ​​five sub-substrates 61 arranged in a row in the X direction. In the example in Figure 4, the group of unit mounting areas 62 arranged in 6 rows in the Y direction is arranged in 3 columns in the X direction. In other words, there are 18 unit mounting areas 62.

[0031] Multiple Fid marks are attached to each unit mounting area 62. Here, we show an example where a first Fid mark FM1 is attached to the -X and -Y corners of each unit mounting area 62, and a second Fid mark FM2 is attached to the +X and +Y corners. The substrate recognition camera 31 captures images of the Fid marks FM1 and FM2, and the positions of these marks are recognized. Based on the results of the position recognition, the movement correction amount for each head 4H is determined. The same procedure is followed for the other unit mounting areas 62. After that, the components that have been attracted to each head 4H are mounted to the mounting area 63 assigned to each head 4H.

[0032] [Control configuration of component mounting device] Figure 5 is a block diagram showing the control configuration of the component mounting device 1. The component mounting device 1 includes a control unit 20 that comprehensively controls the operation of each part of the component mounting device 1. The control unit 20 is electrically connected to the equipment of the head unit 4, camera unit 32U and push-up unit 40, as well as the wafer supply device 6 and component recognition camera 30. When a predetermined program is executed, the control unit 20 operates to functionally comprise an overall control unit 21 (first control unit), axis control unit 22, imaging control unit 23, image processing unit 24 (processing unit / component recognition unit), mounting feasibility determination unit 25 (part of the inspection unit), picking setting unit 26 (second control unit), and storage unit 27.

[0033] The overall control unit 21 comprehensively controls the operation of each functional unit provided by the control unit 20 and performs various calculation processes. For example, the overall control unit 21 performs position recognition of the die 7a and inspection of whether or not components can be mounted on the component mounting points. Position recognition is a process in which the image processing unit 24 recognizes the position of each die 7a on the wafer 7 placed on the wafer stage 10 based on the image acquired by the wafer camera 32. The inspection of whether or not components can be mounted is a process in which the mounting feasibility determination unit 25 determines whether or not components can be mounted on each mounting point of the multi-chamfered substrate PA based on the image acquired by the substrate recognition camera 31 and the measurement results of the laser height measuring instrument 4R. In this embodiment, the overall control unit 21 performs position recognition and the inspection of whether or not components can be mounted in parallel.

[0034] The axis control unit 22 is a driver that drives the servo motors of each unit, and operates each drive motor according to instructions from the overall control unit 21. Specifically, with respect to the head unit 4, the axis control unit 22 controls the movement of the head unit 4 in the XY direction along the ball screw shafts 15 and 19 by controlling the drive of the first Y-axis servo motor 14 and the first X-axis servo motor 18. In addition, the axis control unit 22 controls the movement of the head 4H in the Z direction by controlling the drive of the Z-axis servo motor 401, and controls the rotational movement of the head 4H around its own axis by controlling the drive of the R-axis servo motor 402.

[0035] The axis control unit 22 controls the movement of the camera unit 32U in the XY direction along the ball screw shafts 35 and 39 by controlling the drive of the second Y-axis servo motor 34 and the second X-axis servo motor 38. The axis control unit 22 also controls the movement of the push-up unit 40 in the XY direction along the ball screw shafts 43 and 45 by controlling the drive of the third Y-axis servo motor 44 and the third X-axis servo motor 46. Furthermore, the axis control unit 22 controls the vertical movement of the push-up pin 47 by controlling the drive of the pin lifting motor 48.

[0036] The imaging control unit 23 controls the imaging operations of the component recognition camera 30, the substrate recognition camera 31, and the wafer camera 32. Specifically, the imaging control unit 23 controls the operation of the component recognition camera 30 to image the die 7a or other components that are attracted to the head 4H. The imaging control unit 23 also controls the operation of the substrate recognition camera 31 to image the substrate P or multi-sided substrate PA positioned in the component mounting area of ​​the component mounting device 1. Furthermore, the imaging control unit 23 controls the operation of the wafer camera 32 to image the die 7a in the wafer stage 10.

[0037] The image processing unit 24 performs various image processing operations, including edge extraction, on image data input from the component recognition camera 30, the substrate recognition camera 31, and the wafer camera 32. Based on the image acquired by the component recognition camera 30, the image processing unit 24 performs image processing to recognize whether or not the die 7a or other components are adsorbed onto the adsorption nozzle 52 and their adsorption posture. Based on the image acquired by the substrate recognition camera 31, the image processing unit 24 performs image processing to recognize, for example, the Fid marks FM1, FM2, the mount area 63, and the bad marks 66 of the chamfered substrate PA. Based on the image acquired by the wafer camera 32, the image processing unit 24 performs image processing to recognize the position of the die 7a to be adsorbed on the wafer stage 10.

[0038] The mounting feasibility determination unit 25 performs a process to determine whether a component can be mounted at a predetermined component mounting point on the substrate P or the multi-chamfered substrate PA before the component is mounted. Specifically, the mounting feasibility determination unit 25 determines whether a component can be mounted based on the recognition results of the mounting area 63 and bad marks 66 by the image processing unit 24, and the measurement results of the laser height measuring instrument 4R. If it is recognized that foreign matter or other objects are present at the mounting point of the mounting area 63, or that the bad marks 66 are blacked out, or if the laser height measuring instrument 4R detects an abnormal height in the mounting area 63, the mounting feasibility determination unit 25 determines that the mounting point of the mounting area 63 is "unmountable". If none of the above conditions are detected, the mounting feasibility determination unit 25 determines that the mounting point of the mounting area 63 is "mountable".

[0039] The picking setting unit 26 performs picking setting processing to determine which of the multiple heads 4H will pick each of the multiple dies 7a (parts), and the picking order of the multiple dies 7a by the multiple heads 4H. The picking setting unit 26 performs the picking setting processing based on the position recognition of the dies 7a, which is performed in parallel under the control of the overall control unit 21, and the results of the inspection to determine whether or not the parts can be mounted.

[0040] In a more preferred embodiment, the picking setting unit 26 first performs a provisional picking setting as a first process, assuming that there are no mounting points (mounting areas 63) that are unsuitable for mounting during the component mounting suitability inspection. Subsequently, if the mounting suitability determination unit 25 extracts mounting points that are "unsuitable for mounting" during the component mounting suitability inspection, the picking setting unit 26 performs a second process following the first process, excluding the unsuitable mounting points from the allocation target and performing the picking setting again. The picking setting in the second process is performed before the head unit 4 actually picks the die 7a.

[0041] The storage unit 27 stores various programs and data, such as implementation programs. In this embodiment, the storage unit 27 stores array data such as the array pitch of the heads 4H of the head unit 4, the array pitch of the dies 7a on the wafer stage 10, and the array pitch of the mounting points on the mounting area 63 of the chamfered substrate PA. Furthermore, the storage unit 27 stores the amount of movement of the head unit 4, which is determined in advance based on this array data. This amount of movement includes a first amount of movement of the head unit 4 when mounting components, in which multiple heads 4H each mount a die 7a on the mounting area 63, and a second amount of movement of the head unit 4 when picking, in which multiple heads 4H each pick up a die 7a.

[0042] [Component mounting flow] Next, the flow of the component mounting operation by the component mounting device 1 will be explained based on Figures 6 and 7. First, referring to Figure 6, the flow of a comparative example corresponding to the prior art of this embodiment will be explained. The "first suction group" and "second suction group" described in Figures 6 and 7 refer to groups of components (a total of 10 dies 7a) picked up in one suction operation by the head unit 4, which has 10 heads 4H, as illustrated in Figure 4. Component mounting is performed in units of "first suction group" and "second suction group". The operation of mounting components picked by one suction group onto one unit mounting area 62 of the multi-chamfered substrate PA is repeated.

[0043] In the comparative example, first, for the "first adsorption group," a pre-mounting inspection process is performed to detect mounting points where components cannot be mounted based on the image of the multi-sided substrate PA acquired by the substrate recognition camera 31 (step S1). Next, a component recognition process is performed to recognize the position of the die 7a based on the image of the wafer 7 acquired by the wafer camera 32 (step S2). After that, the assignment of each head 4H of the head unit 4 to be picked and the die 7a to be picked is determined, which corresponds to the picking setting described above (step S3).

[0044] Subsequently, component suction is performed in which the dies 7a designated as the "first suction group" are suctioned onto the suction nozzles 52 of each head 4H according to the aforementioned allocation (step S4). Furthermore, component mounting is performed in which the head unit 4 is moved onto the multi-sided substrate PA and the suctioned dies 7a are mounted onto one unit mounting area 62 (step S5). After that, steps S6 to S10, similar to steps S1 to S5 above, are performed for the "second suction group". The same applies to the third suction group and subsequent groups.

[0045] In the comparative example, after performing the pre-mounting inspection in step S1, the component recognition in step S2 and the picking setting including allocation in step S3 are performed. Therefore, on the multi-sided substrate PA, the picking setting can be performed after identifying mounting points that are "unmountable". Thus, the process can be simplified. However, there may be a waiting time for the processing of step S1 when executing steps S2 and S3, which may worsen the cycle time. Also, the fact that the mounting process for the "second suction group" starts after the mounting process for the "first suction group" is completed is undesirable from the standpoint of improving the cycle time.

[0046] Figure 7 is a flowchart illustrating a component mounting operation according to an embodiment to which the present invention is applied. Figure 7 includes a time axis t, and steps performed in parallel are listed side by side on the time axis t. In the embodiment, the overall control unit 21 causes the "first suction group" to perform the pre-substrate mounting inspection process (step S11) and the component recognition process (step S12) in parallel in time.

[0047] In step S11, the pre-mounting inspection of the substrate, the image processing unit 24 performs predetermined image processing on the image of the multi-sided substrate PA acquired by the substrate recognition camera 31. Based on the results of this image processing, the mounting feasibility determination unit 25 determines whether a predetermined component mounting point is "mountable" or "unmountable". In step S12, the component recognition, the image processing unit 24 performs predetermined image processing on the image of the wafer 7 acquired by the wafer camera 32 to identify the coordinate position of the die 7a to be picked up. Following step S12, as a first process, the picking setting unit 26 provisionally performs the picking settings described above, assuming that no "unmountable" component mounting points are detected in the inspection of step S11, which is performed in parallel. That is, it provisionally determines the assignment of each head 4H of the head unit 4 to the die 7a to be picked (step S13).

[0048] Next, based on the inspection results in step S11, the picking setting unit 26 performs the picking settings again as a second process (step S14). If no "unmountable" component mounting points are detected in step S11, the provisional picking settings determined in step S13 are used as is. On the other hand, if "unmountable" component mounting points are extracted in step S11, the picking setting unit 26 removes the "unmountable" component mounting points from the allocation target and redoes the picking settings. Specific examples of the re-picking settings will be described in detail later, referring to Figures 8 to 13C. The above picking settings are performed before the head unit 4 picks the die 7a.

[0049] Subsequently, according to the picking settings in step S14, the axis control unit 22 controls the operation of the head unit 4 to perform component picking, picking up the dies 7a designated as the "first picking group" onto the picking nozzles 52 of each head 4H (step S15). Furthermore, the axis control unit 22 moves the head unit 4 onto the chamfered substrate PA and performs component mounting, placing the 10 picked-up dies 7a into a single unit mounting area 62 (step S16).

[0050] In parallel with step S16, the component recognition process for the "second picking group" (step S22) is performed. This is because the head unit 4 moves from the component supply unit 5 onto the chamfered substrate PA, making the wafer stage 10 available and enabling the wafer camera 32 to image the wafer 7. Based on the recognition results of step S22, the picking setting unit 26 tentatively determines the assignment of each head 4H to the die 7a (step S23).

[0051] Meanwhile, in parallel with the processing in steps S22 and S23, a pre-mounting inspection of the "second suction group" is performed (step S21). That is, once the component mounting in step S16 of the "first suction group" is completed, the mounting feasibility determination unit 25 determines whether or not components can be mounted at the mounting points of the "second suction group". After that, in the same manner as in steps S14, S15, and S16, the final determination of the allocation between each head 4H and die 7a (step S24), component suction (step S25), and component mounting (step S26) are performed for the "second suction group". After the completion of step S26, a pre-mounting inspection of the "third suction group" is performed (step S31). Furthermore, in parallel with step S26, the component recognition process for the "third suction group" (step S32) is performed. The following processes are the same as described above.

[0052] According to the above embodiment, the cycle time can be improved in a component mounting apparatus 1 that directly picks dies 7a from wafers 7 using multiple heads 4H of the head unit 4. Specifically, instead of sequentially performing die 7a position recognition and substrate P mounting point feasibility checks, these are performed in parallel, thus shortening the cycle time. Furthermore, since the picking setting unit 26 performs picking settings after die 7a position recognition and substrate mounting inspection, accurate allocation between each head 4H, die 7a, and mounting points, as well as setting the picking order, can be achieved. In addition, if no mounting points that are "unmountable" are detected, the temporary picking settings from steps S13 and S23 can be used as is. Therefore, picking settings that support both detected and undetected mounting points can be easily implemented.

[0053] [Specific examples of picking settings] Next, a specific example of picking settings by the picking setting unit 26 will be explained. First, referring to Figure 8, the numbering assignments for the component mounting points of the multi-sided substrate PA, each head 4H of the head unit 4, and the die 7a of the wafer 7 will be explained. The 10 heads 4H of the head unit 4 are each assigned head numbers from 1 to 10.

[0054] The component mounting points on the substrate PA are numbered 1 to 10, indicating that each of the heads 4H numbered 1 to 10 is the position where the die 7a will be mounted. 1G, 2G, and 3G indicate the rows where the dies 7a picked up by the first, second, and third pick-up groups described above will be mounted, respectively. Each die 7a on the wafer 7 is also numbered 1 to 10, indicating that it will be picked up by each of the heads 4H numbered 1 to 10. The XY symbols attached to the wafer 7 are directional indicators corresponding to the XY directions shown in Figure 3A, and each XY number is an XY coordinate value indicating the address of each die 7a.

[0055] For example, die 7a located at the (X15, Y4) coordinates of wafer 7 is assigned the number 10. This die 7a, number 10, is picked up by head 4H, which is also assigned the number 10. In the case of the first picking group, die 7a, picked up by head 4H, is mounted at mounting point 10 on substrate PA at [1G]. Here, the heads 4H are numbered from 10 to 1, from left to right in Figure 8. When picking to the right in Figure 8, the heads 4H pick in the order of 10, 9, and 8. On the other hand, when picking to the left, the heads 4H pick in the order of 1, 2, and 3. By setting it this way, the amount of movement of the head unit 4 can be suppressed. In the example shown in Figure 8, picking of die 7a proceeds from head 10 to head 7 4H in the order of (X15~18, Y4), then a new line is created and picking of die 7a proceeds from head 1 to head 6 4H in the order of (X20~15, Y3).

[0056] The following shows specific examples of how to change the picking settings of head 4H when a mounting point deemed "unmountable" (hereinafter sometimes referred to as "mounting point NG") is identified during the pre-mounting inspection of the substrate. Here, we illustrate the mount priority mode (second mode), which minimizes the movement of head unit 4 when mounting components onto a multi-sided substrate PA, and the picking priority mode (first mode), which minimizes the movement of head unit 4 when picking up die 7a from wafer 7.

[0057] <Mount Priority Mode> Figures 9 and 10 are schematic diagrams illustrating an example of changing the picking settings in mount priority mode when a mounting point failure occurs. Figure 9 assumes that mounting point number 8 of the first suction group [1G] of the multi-sided substrate PA is extracted as a mounting point failure. In the provisional picking settings (step S13 in Figure 7), head number 8 4H is assigned to pick up die number 8 7a at address (X17, Y4) of wafer 7 and mount die number 8 7a.

[0058] Figure 10 is a schematic diagram illustrating the re-execution of the picking settings in mount-priority mode. In mount-priority mode, the assignment is modified so that die 7a is picked, skipping head 4H, which was assigned to mounting point NG. Specifically, die 7a (target part) at address (X17, Y4), which was assigned to mounting point NG, is assigned to head 4H, which is adjacent to head 4H in the suction order, instead of head 4H (target head), which was assigned in the temporary setting. In other words, die 7a at (X17, Y4) is picked up by head 4H, not head 4H.

[0059] In this way, the picking setting unit 26 performs a new picking setting (step S14 in Figure 7) so that the die 7a that was assigned to mounting point NG is not left on the wafer 7 of the wafer stage 10. That is, the die 7a at (X17, Y4) is picked up by the head 4H without being left on the wafer 7. Furthermore, the dies 7a to be picked up by the remaining heads 6 to 1 4H are also shifted one by one from the provisional setting. Specifically, the die 7a at (X18, Y4) is changed from head 7 to head 6 4H. For the die 7a in row Y3, the assignment to heads 1 to 5 4H is maintained as in the provisional setting. This is because, as shown in the pickup route R in Figure 8, the picking for row Y3 is to the left.

[0060] If the system were to leave die 7a (X17, Y4) assigned to mounting point NG in the provisional settings, then an operation to pick that die 7a would be required later. In this case, extra cycle time may be generated, and combined with the fact that surrounding die 7a are being picked, it may become impossible to accurately recognize and pick the part. This embodiment avoids such problems.

[0061] With the picking settings changed as described above, the head unit 4 completes the picking operation of the die 7a of the first picking group [1G] from the wafer 7, with only head 8H not picking up die 7a. Then, it moves above the multi-sided substrate PA, and the die 7a picked up by each head 4H are sequentially mounted, except for mounting point 8, which is mounting point NG.

[0062] In the mount-priority mode described above, the aim is to make the amount of movement of the head unit 4 during component mounting smaller than the amount of movement of the head unit 4 during picking, when each head 4H picks up a die 7a. Therefore, the cycle time, mainly during component mounting, can be improved. This point will be explained in detail with reference to Figures 11A to 11C.

[0063] Figures 11A to 11C illustrate specific examples of component mounting operation using mount priority mode. Here, we assume that mounting point 8 of the multi-chamfered substrate PA is a mounting point NG, and that the picking settings are re-executed as shown in Figure 10. Step #11 in Figure 11A shows the state where head 4H picks die 7a 10 on wafer 7, and then head 4H picks die 7a 9. Since no component has been assigned to head 4H, no picking is performed by head 4H.

[0064] Step #12 in Figure 11A shows the state in which head 4H (number 7) is being aligned with die 7a (number 7) located at (X17, Y4) on wafer 7. In the state of step #11, head 4H (number 8) is closest to die 7a (X17, Y4) to be picked next. However, in this mode, head 4H (number 8) is skipped, and head 4H (number 7) is made to pick die 7a (X17, Y4). Therefore, in order to perform the alignment, it is necessary to move the head unit 4 by a relatively large amount of movement Δm1.

[0065] Step #13 in Figure 11B shows the state where head 4H (number 7) has picked die 7a (number 7). Head 4H (number 8) does not hold die 7a. Subsequently, along the pickup route R, heads 6 (number 6), 1 (number 1), 2 (number 2), and so on, picking of the assigned die 7a is performed. After picking is complete, head unit 4 is moved above the chamfered substrate PA, and the mounting of the picked die 7a onto the chamfered substrate PA begins. Step #14 in Figure 11B shows the state where die 7a has been mounted from head 4H (number 10) to mounting point 10 on the chamfered substrate PA, and from head 4H (number 9) to mounting point 9. Mounting point 8 is not a mounting point (NG), so die 7a is not mounted there.

[0066] Step #15 in Figure 11C shows the state where head 4H (number 7) is aligned to mounting point 7. In this alignment, the head unit 4 is aligned to the mounting point originally assigned to head 4H (number 7), so the amount of movement of the head unit 4 is relatively small, Δm2. In other words, Δm1 > Δm2 can be made, so the cycle time during component mounting can be shortened in the event of a mounting point failure. Step #16 in Figure 11C shows the state where die 7a has been mounted from head 4H (number 7) to mounting point 7. Subsequently, die 7a is mounted from heads 4H (numbers 6-1) to mounting points 6-1, respectively, completing component mounting for the first suction group [1G]. Next, picking and component mounting for the second suction group [2G] begins.

[0067] <Picking priority mode> Figure 12 is a schematic diagram showing how the picking settings are re-executed in picking priority mode. In picking priority mode, the assignment is modified so that head 4H, which was assigned to mounting point NG, picks die 7a, skipping mounting point NG and mounting the die.

[0068] Specifically, die 7a (target component) at address (X17, Y4), which was assigned to mounting point 8 (NG), will be assigned to head 4H (target head) at mounting point 8, as per the provisional assignment. Then, head 4H at mounting point 1 will be removed from the picking target as it will not be mounted at mounting point 8. In other words, die 7a at (X20, Y3) will be reassigned from head 1 to head 2. Die 7a at (X19~16, Y3) will also be reassigned so that the head number is shifted by one. Meanwhile, on the multi-sided board PA side, head 4H at mounting point 8 will be assigned to mounting point 7. Heads 4H at mounting points 7 through 2 will also be reassigned so that their head numbers are shifted by one, to mounting points 6 through 1 respectively.

[0069] With the picking settings changed as described above, head unit 4 completes the picking operation of the die 7a of the first picking group [1G] from wafer 7, with only head 1 4H not picking up die 7a. Then, it moves above the multi-sided substrate PA, and the die 7a picked up by each head 4H are sequentially mounted, except for mounting point 8, which is mounting point NG.

[0070] In the picking priority mode described above, the aim is to make the movement of the head unit 4 during picking of die 7a from wafer 7 smaller than the movement of the head unit 4 during component mounting by each head 4H. Therefore, the cycle time during picking can be improved. This point will be explained in detail with reference to Figures 13A to 13C.

[0071] Figures 13A to 13C illustrate specific examples of component mounting operations in picking priority mode. Here, we assume that mounting point 8 of the multi-chamfered substrate PA is a mounting point NG, and that the picking settings are re-executed as shown in Figure 10. Step #21 in Figure 13A shows the state where head 4H picks die 7a at position 10, followed by head 4H picking die 7a at position 9. In picking priority mode, die 7a at position 8 is also assigned to head 4H at position 8, so the next picking operation is performed by head 4H at position 8.

[0072] Step #22 in Figure 13A shows the state in which head #8 4H is being aligned with die #8 7a located at (X17, Y4) on wafer 7. In the state of step #21, head #8 4H is closest to die 7a at (X17, Y4) to be picked next. Therefore, the amount of movement of the head unit 4 for the alignment is a relatively small amount Δm3.

[0073] Step #23 in Figure 13B shows the state where head 4H (number 8) has picked die 7a (number 8). Subsequently, heads 7, 2, 3, and so on pick their assigned dies 7a. Head 4H (number 1) does not pick die 7a. After picking is complete, head unit 4 is moved above the chamfered substrate PA, and the mounting of the picked dies 7a onto the chamfered substrate PA begins. Step #24 in Figure 13B shows the state where die 7a has been mounted from head 4H (number 10) to mounting point 10 on the chamfered substrate PA, and from head 4H (number 9) to mounting point 9. Mounting point 8 is not a mounting point, so die 7a is not mounted there.

[0074] Step #25 in Figure 13C shows the state in which head #8 4H is positioned at mounting point #7. In this positioning, head #8 4H is shifted to the mounting point previously assigned to the adjacent head #7 4H, so a relatively large movement amount Δm4 of head unit 4 is required. In other words, Δm3 < Δm4, so the takt time during picking can be shortened if a mounting point error occurs.

[0075] Step #26 in Figure 13C shows the state where die 7a has been mounted from head 8H to mounting point 7. Subsequently, die 7a is mounted from heads 7-2H to mounting points 6-1, completing the component mounting for the first suction group [1G]. Next, picking and component mounting for the second suction group [2G] begins.

[0076] [How to choose between the two component mounting modes] The two component mounting modes described above, mount-priority mode and picking-priority mode, can be said to have a trade-off relationship in terms of improving cycle time. The picking setting unit 26 needs to select one of the modes when mounting points NG are detected on the substrate. The desirable selection method is to consider the picking process and the component mounting process as a whole and select the mode that has the best cycle time. Which mode is superior in terms of cycle time depends on the relationship between the pitch of the die 7a on the wafer stage 10 or the pitch of the mounting points on the substrate PA and the arrangement pitch of the heads 4H on the head unit 4. An example of calculating which mode is superior in terms of cycle time by considering this relationship and adopting the dominant mode is shown below.

[0077] Figure 14 is a schematic diagram showing the relationship between the pitch of head 4H, the arrangement pitch or component mounting point pitch of die 7a, and the amount of movement of head 4H. Similar to the example above, we assume a case where the mounting point assigned to head 4H (number 8) is determined to be an NG mounting point. The following patterns (1) to (3) are possible for the pitch magnitude relationship. • Pattern (1): Pitch of die 7a or board mounting point > Pitch of head 4H • Pattern (2): Pitch of head 4H > Pitch of die 7a or mounting point on the board • Pattern (3): Pitch of die 7a or board mounting point > Pitch of head 4H × 2

[0078] Figure 14 schematically shows heads 9, 8, and 7 of head unit 4, with the head pitch represented by "H". Additionally, adjacent die 7a on wafer 7 or adjacent substrate mounting points on the chamfered substrate PA are indicated by "X" and "Y", with their pitch represented by "Dn" (n=1~3).

[0079] Pattern (1) is a pattern in which the pitch D1 of "X" and "Y" is greater than the head pitch H. In Pattern (1), the amount of movement of head 4H in Pattern (1)-1, which corresponds to the picking priority mode (first mode) in which the component (die 7a) is attracted and mounted on head 8, and in Pattern (1)-2, which corresponds to the mount priority mode (second mode) in which head 8 is skipped and the component is attracted and mounted on head 7, is as follows. In Pattern (1)-1, the amount of movement a1 required to move head 8 to the center of "Y" (the center of die 7a or the center of the mounting point) is: a1 = D1 - H This is the result. On the other hand, in pattern (1)-2, the amount of movement b1 required to move the 7-iron head to the center of "Y" is, b1 = 2 × H - D1 This is the result.

[0080] Next, in pattern (2)-1, which corresponds to the picking priority mode in pattern (2) where a component is attracted to and mounted on the 8th head, the amount of movement a2 required to move the 8th head to the center of "Y" is: a2 = H - D2 This is the result. In pattern (2)-2, which corresponds to the mount priority mode in pattern (2) where a component is attracted and mounted to the 7th head, the amount of movement b2 required to move the 7th head to the center of "Y" is: b² = 2 × H - D² This is the result.

[0081] Furthermore, in pattern (3)-1, which corresponds to the picking priority mode in pattern (3) where a component is attracted to and mounted on the 8th head, the amount of movement a3 required to move the 8th head to the center of "Y" is: a3 = D3 - H In pattern (3)-2, which corresponds to the mount priority mode in pattern (2) where a component is attracted and mounted to the 7th head, the amount of movement b3 required to move the 7th head to the center of "Y" is: b3 = D3 - 2 × H This is the result.

[0082] Figure 15 is a tabular diagram showing the formula for calculating the movement amount of the head unit 4 in each pattern when the picking priority mode is selected. Two cases are assumed for "X" and "Y" above: when they are the die 7a, and when they are the mounting points on the circuit board, and the relationship between the head pitch H and these cases is assumed. There are three patterns in the picking priority mode, (1)-1, (1)-2, and (1)-3 in Figure 14, so as shown in Figure 15, there are nine combinations of the relationship between the head pitch and Y, from (A) to (I). In Figure 15, the distances when "X" and "Y" are the die 7a (Pick) are written as a1, a2, a3, and the die pitch is written as D1, D2, D3, as in Figure 14. To distinguish this, the distances when "X" and "Y" are the mounting points on the circuit board (Mount) are written as a1p, a2p, a3p, and the mounting point pitch is written as D1p, D2p, D3p, with a lowercase "p" added.

[0083] Figure 16 is a tabular diagram showing the formulas for calculating the amount of movement of the head unit 4 in each pattern when the mount priority mode is selected. The mount priority mode also has three patterns, (2)-1, (2)-2, and (2)-3 as shown in Figure 14, so as shown in Figure 15, there are nine conditions (A) to (I) for the combination of magnitude relationships. In Figure 15, the distances when "X" and "Y" are the die 7a (Pick) are denoted as b1, b2, b3, and the die pitches are denoted as D1, D2, and D3, as in Figure 14. To distinguish this, the distances when "X" and "Y" are the board mounting points (Mount) are denoted as b1p, b2p, b3p, and the mounting point pitches are denoted as D1p, D2p, and D3p.

[0084] The movement amounts a1, a2, a3 and b1, b2, b3 (hereinafter referred to as the first movement amount) of the head unit 4 during picking, and the movement amounts a1p, a2p, a3p and b1p, b2p, b3p (hereinafter referred to as the second movement amount) of the head unit 4 during component mounting are stored in advance in the storage unit 27 (Figure 5). That is, for the substrate P and components scheduled to be produced in the component mounting device 1, the values ​​of the substrate mounting point pitch, component pitch, and head pitch, along with the values ​​of the movement amounts of the head unit 4 mentioned above, are stored in the storage unit 27.

[0085] The picking setting unit 26 (second control unit) refers to the first and second movement amounts described above and sets the picking settings by adopting whichever of the picking priority mode or mounting priority mode is more advantageous in terms of cycle time. Specifically, the picking setting unit 26 calculates the first sum of the first and second movement amounts when the picking priority mode is adopted, and the second sum of the first and second movement amounts when the mounting priority mode is adopted. Then, the picking setting unit 26 compares the first sum and the second sum and sets the picking settings in such a manner that it selects the picking priority mode if the first sum is smaller, and selects the mounting priority mode if the second sum is smaller.

[0086] For example, consider the case of condition (A) in Figures 15 and 16. The picking setting unit 26 calculates a1 + a1p as the first total and b1 + b1p as the second total, and then calculates the difference DF between the two. That is, DF = (a1 + a1p) - (b1 + b1p) =(D1-H+D1p-H)-(2×H-D1+2×H-D1p) =2D1+2D1p-6H The differential DF is calculated. If the differential DF is positive, then (a1 + a1p) is larger, meaning the total movement of head unit 4 is larger, which is disadvantageous in terms of cycle time. Therefore, the mount-priority mode, which causes the component to be attracted and mounted to head 7, is selected. Conversely, if the differential DF is negative, the picking-priority mode, which causes the component to be attracted and mounted to head 8, is selected. According to this embodiment, it is possible to perform component mounting in a mode that can further shorten the cycle time, taking into account both component mounting and picking.

[0087] [Differentiation] Although embodiments of the present invention have been described above, the present invention is not limited to the embodiments described above. For example, the following modified embodiments can be adopted.

[0088] (1) In the above embodiment, the explanation was based on an example in which a mounting point NG was detected only at mounting point No. 8 of the multi-sided substrate PA, as shown in the condition settings in Figure 9. By applying the example of a single mounting point NG, the picking settings can be changed in the same way when two or more mounting point NGs are detected per suction group.

[0089] (2) In the above embodiment, the multi-chamfered substrate PA shown in Figure 4 was used as an example substrate. The multi-chamfered substrate PA is just one example of a "substrate" according to the present invention, and other types of printed circuit boards other than the multi-chamfered substrate PA may also be used.

[0090] (3) In the above embodiment, the amount of movement of the head unit 4 during picking and component mounting was described solely as an evaluation criterion for the quality of the cycle time. The cycle time is also affected by the amount of movement of the head unit 4 from the component picking position to the component mounting position, and the amount of movement of the head unit 4 for picking the next group of components. Therefore, when changing the picking settings, it is desirable to also consider the amount of movement of the head unit 4 at times other than during picking and component mounting.

[0091] [Inventions included in the above embodiments] The specific embodiments described above mainly include the invention having the following configuration.

[0092] A component mounting apparatus according to one aspect of the present invention comprises: a component supply unit having a component placement area on which a plurality of components arranged in a matrix are arranged; a component recognition unit having a camera for imaging the components in the component placement area and a processing unit that recognizes the position of the components in the component placement area based on the image acquired by the camera; a head unit having a plurality of heads that can move between a component mounting area on which a substrate on which the components are mounted is arranged and the component placement area, and picks the components in the component placement area and mounts the components onto the substrate in the component mounting area; an inspection unit that checks whether components can be mounted on predetermined mounting points on the substrate arranged in the component mounting area; a first control unit that controls the operation of the component recognition unit and the inspection unit; and a second control unit that executes a picking setting process that determines which of the plurality of heads each of the plurality of components is to be picked, and the picking order of the plurality of components by the plurality of heads. The first control unit causes the component recognition unit to perform position recognition of the components and the inspection unit to perform the mounting feasibility check in parallel. The second control unit executes the picking setting process based on the results of the position recognition of the component and the inspection of whether or not it can be mounted, which are performed in parallel.

[0093] This component mounting device improves the cycle time in a component mounting device that directly picks multiple components from a component placement area using multiple heads on a head unit. Specifically, instead of sequentially performing component position recognition and board mounting point inspection, these are performed in parallel, thus shortening the cycle time. Furthermore, since picking settings are performed after the position recognition and inspection, accurate allocation and picking order settings can be achieved.

[0094] In the component mounting device described above, the second control unit may, as a first process, assume that there are no mounting points that cannot be mounted in the mounting feasibility inspection and provisionally perform the picking settings. If mounting points that cannot be mounted are identified in the mounting feasibility inspection, the second control unit may, as a second process following the first process, remove the mounting points that cannot be mounted from the allocation target and perform the picking settings again.

[0095] In this embodiment, a provisional picking setting is made in the first process, and if the inspection unit detects a mounting point that cannot be mounted, a definitive picking setting is made in the second process based on this. If no mounting points that cannot be mounted are detected, the provisional picking setting can be used as is. Therefore, picking settings that can handle both cases, with and without the detection of mounting points that cannot be mounted, can be easily made.

[0096] In the above-described component mounting device, it is desirable that the second control unit perform the picking settings in the second process before the head unit performs the picking.

[0097] According to this embodiment, the timing of picking settings can be optimized.

[0098] In the component mounting device described above, it is desirable that the second control unit perform the picking settings in the second process so that components that were assigned to mounting points where mounting is not possible in the first process do not remain in the component placement area.

[0099] If the system leaves behind components assigned to mounting points where mounting is not possible, a picking operation for those components will be required later. In this case, extra cycle time may be generated, and coupled with the fact that surrounding components have been picked, it may become difficult to accurately recognize and pick the components. Such problems can be resolved by the above-described embodiment.

[0100] In the component mounting apparatus described above, the second control unit may configure the picking settings in the second process such that the amount of movement of the head unit during component mounting, when the plurality of heads each mount the component onto the substrate, is smaller than the amount of movement of the head unit during picking, when the plurality of heads each pick up the component.

[0101] In this embodiment, priority is given to minimizing the amount of movement of the head unit during component mounting, and the picking settings in the second process are performed accordingly. Therefore, the cycle time, mainly during component mounting, can be improved.

[0102] In the component mounting apparatus described above, the second control unit may configure the picking settings in the second process such that the amount of movement of the head unit during picking, in which the plurality of heads each pick up the component, is smaller than the amount of movement of the head unit during component mounting, in which the plurality of heads each mount the component onto the substrate.

[0103] In this configuration, priority is given to minimizing the movement of the head unit during picking, and the picking settings are performed in the second process. Therefore, the cycle time during picking can be improved.

[0104] The above-described component mounting device includes a storage unit that stores, in advance, the first movement amount of the head unit during component mounting, in which the multiple heads each mount the component on the substrate, and the second movement amount of the head unit during picking, in which the multiple heads each pick up the component, based on the arrangement pitch of the multiple heads and the arrangement pitch of the multiple components in the component placement area or the arrangement pitch of the mounting points on the substrate. The second control unit, in setting the picking in the second process, determines the target component that was assigned to the mounting point where mounting is not possible in the first process, and picks the target component. In the first mode, which involves selecting a first mode in which the target head assigned in the first process picks the target component and mounts it onto the substrate, a first total is calculated, which is the sum of the first and second movement amounts. In the second mode, which involves selecting a second mode in which the target component is picked by an adjacent head adjacent to the target head and mounted onto the substrate, a second total is calculated, which is the sum of the first and second movement amounts. The first total and the second total are then compared, and if the first total is smaller, the first mode is selected. If the second total is smaller, the second mode is selected. This configuration may also be used to perform the picking settings.

[0105] According to this embodiment, the method for picking components assigned to mounting points that cannot be mounted is determined by comparing the first total and the second total, and whether to pick them using the target head (first mode) or the adjacent head (second mode). The first total and the second total are the sum of the movement amounts of the head unit during component mounting and picking. Therefore, it is possible to mount components in a mode that can shorten the cycle time by considering both component mounting and picking.

[0106] In the component mounting apparatus described above, the component placement area is an area where a diced wafer is placed, the component is a die, the camera is a wafer camera for imaging the wafer, and it is preferable that the plurality of heads are arranged in a straight line.

[0107] According to this embodiment, the cycle time can be improved in a component mounting apparatus that picks dies from a diced wafer and mounts the dies onto a substrate.

Claims

1. A parts supply unit having a parts placement area where multiple parts arranged in a matrix are placed, A component recognition unit having a camera that captures images of the component in the component placement area, and a processing unit that recognizes the position of the component in the component placement area based on the image acquired by the camera, A head unit having multiple heads that can move between a component mounting area where a circuit board on which the aforementioned components are mounted is located and a component placement area, and which picks the aforementioned components in the component placement area and mounts the aforementioned components onto the circuit board in the component mounting area, With respect to the circuit board arranged in the component mounting area, an inspection unit is provided to check whether components can be mounted at predetermined mounting points, A first control unit that controls the operation of the component recognition unit and the inspection unit, The system includes a second control unit that performs a picking setting process which determines which of the multiple heads will pick each of the multiple parts, and the picking order of the multiple parts by the multiple heads. The first control unit performs the position recognition of the component by the component recognition unit and the inspection of whether or not it can be mounted by the inspection unit in parallel. In a component mounting device, the second control unit performs the picking setting process based on the results of the position recognition and mounting feasibility inspection of the component, which are performed in parallel. As a first process, the second control unit assumes that there are no mounting points that are unsuitable for mounting in the mounting suitability inspection, and provisionally sets the picking configuration. If, in the inspection of whether or not a mounting point can be mounted, an unsuitable mounting point is identified, the second control unit, as a second process following the first process, removes the unsuitable mounting point from the allocation target and performs the picking setting again, the component mounting device.

2. In the component mounting device according to Claim 1, The component mounting device wherein the second control unit performs the picking settings in the second process before the head unit performs the picking.

3. In the component mounting device according to Claim 1, The second control unit performs the picking settings in the second process so that components that were assigned to mounting points where mounting is not possible in the first process do not remain in the component placement area.

4. In the component mounting device described in Claim 3, A component mounting apparatus, wherein the second control unit sets the picking settings in the second process such that the amount of movement of the head unit during component mounting, when the plurality of heads each mount the component onto the substrate, is smaller than the amount of movement of the head unit during picking, when the plurality of heads each pick up the component.

5. In the component mounting device according to claim 3, A component mounting apparatus, wherein the second control unit sets the picking settings in the second process such that the amount of movement of the head unit during picking, in which the plurality of heads each pick up the components, is smaller than the amount of movement of the head unit during component mounting, in which the plurality of heads each mount the components onto the substrate.

6. In the component mounting device according to claim 3, The system includes a storage unit that stores, in advance, the first amount of movement of the head unit during component mounting, in which the multiple heads each mount the components onto the substrate, and the second amount of movement of the head unit during picking, in which the multiple heads each pick up the components, based on the array pitch of the multiple heads and the array pitch of the multiple components in the component placement area or the array pitch of the mounting points on the substrate. The second control unit, in setting the picking in the second process, In the first process, the target component that was assigned to the mounting point where mounting is not possible is picked by the target head that was assigned in the first process to pick the target component, and the target component is mounted on the substrate. In the first mode selected, the first total is the sum of the first movement amount and the second movement amount, When a second mode is selected in which the target component is picked by an adjacent head adjacent to the target head and the target component is mounted on the substrate, the second total, which is the sum of the first movement amount and the second movement amount, is calculated. A component mounting device that compares the first total with the second total, and selects the first mode if the first total is smaller, and selects the second mode if the second total is smaller, to perform the picking settings.

7. In the component mounting device according to any one of claims 1 to 6, The aforementioned component placement area is an area where a diced wafer is placed, and the component is a die. The camera is a wafer camera that images the wafer, The aforementioned multiple heads are arranged in a straight line in a component mounting device.

Images