JP2010199630A - Method and apparatus for mounting electronic component - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily and exactly determine a correction value of a rotation center position of each spindle of a rotary head of an exchanged component transfer apparatus in an electronic component mounting apparatus in which the component transfer apparatus equipped with a rotary head is exchangeably fixed to a transport table. <P>SOLUTION: The transport table is transported such that the rotation center of a spindle indexed on a mounting point among a plurality of spindles of the rotary head is positioned on a coordinate position of an optical axis of a calibrated camera for recognition of components. Images of the tip of a suction nozzle installed on each spindle are taken by the camera for recognition of components in the state that the each spindle is positioned on a first rotation angular position and a second rotation angular position different from the first rotation angular position, respectively. Based on the center position of the image of the tip of the each suction nozzle on the first and second angular positions indexed on the mounting point, each correction value of the rotation center position of the each spindle can efficiently be determined. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a calibration method in an electronic component mounting apparatus for mounting electronic components on a substrate, and in particular, a calibration method for preventing an error in the mounting position of a component when a component transfer apparatus having a rotary head is replaced. About.

  Electronic components that are mounted on a board by this type of electronic component mounting device are small electronic components that are used in large quantities, such as chips. There are large electronic components such as ICs with fine leads that are used. Normally, small electronic components that are used in large numbers are mounted with a mounting device for high-speed small components that has multiple suction nozzles, while large electronic components are relatively slow, but they can handle a wide variety of components and have high-precision mounting functions. It is mounted by the mounting apparatus for odd-shaped electronic components provided with. For this reason, conventional electronic component mounting lines are often formed by arranging high-speed small-component mounting devices and odd-shaped electronic component mounting devices in series. Alternatively, an electronic component mounting line is formed by using a plurality of electronic component mounting apparatuses having both a high-speed small component mounting function and an irregular large electronic component mounting function in one model.

  However, since the mounting ratio of small electronic components and large electronic components on the board differs depending on the type of board, the former method of forming an electronic component mounting line by combining a mounting device for high-speed small components and a mounting device for odd-shaped electronic components, There is a problem in that it is impossible to construct an electronic component mounting line that always obtains the maximum productivity because the operation rate of either one of the electronic component mounting apparatuses must be reduced. In the latter method using an electronic component mounting apparatus having both a high-speed small component mounting function and an irregular large electronic component mounting function, each electronic component mounting apparatus depends on whether the component to be mounted is a small electronic component or a large electronic component. However, there was a problem that some of the functions became excessive more than necessary, and the equipment cost rebounded to the component mounting cost. Furthermore, since the types of components mounted on the board are steadily changing and it is difficult to predict the types at the time of designing the electronic component mounting apparatus, the electronic component mounting apparatus can cope with the types of components beyond the current level. Necessitates a new design, which necessitates surplus equipment and increases equipment costs.

  On the other hand, in the electronic component mounting apparatus, the component transfer device that holds and mounts the electronic component can be exchanged so that an operator can replace the component transfer device in a short time at the site of the component mounting line. Thus, there has been proposed a method of constructing an optimal component mounting line according to the ratio of small electronic components and large electronic components on the board to be produced and the type of component.

  In this type of electronic component mounting apparatus, for example, as shown in FIG. 5, a component transfer apparatus 26 having a component mounting head 28 on a movable table 24 supported so as to be movable in two directions X and Y with respect to the base 11. The board recognition camera 25 is provided, and the component recognition camera 15 is fixed to the base 11. Then, the electronic component mounting apparatus 10 detects the position of the board mark Sm provided on the board S that has been carried and positioned by the board conveying apparatus 12 by the board recognition camera 25, and based on the position of the board mark Sm. The position P is corrected, the slide 21 and the moving table 24 are moved in the X direction and the Y direction, and the component P sucked to the tip of the suction nozzle 29 of the component mounting head 28 from the component supply device 13 is set to predetermined coordinates on the substrate S. Implemented in position. Further, while the component P sucked at the tip of the suction nozzle 29 is moved from the component supply device 13 to a predetermined coordinate position on the substrate S, the suction nozzle 29 is temporarily stopped by the component recognition camera 15 and the center of the suction nozzle 29 is stopped. The component misalignment of the component P with respect to the line O3 (hereinafter also referred to as the suction nozzle centerline O3) is detected by the component recognition camera 15, and this also corrects the movement amount of the slide 21 and the moving table 24, so that the component P It is designed to be mounted accurately at the coordinate position on the substrate S.

  In the case where a plurality of different types of component transfer devices are exchanged for the mounting device basic portion as described above, in order to accurately mount the component P at a predetermined coordinate position on the substrate S, FIG. In FIG. 1, which is a partially enlarged view of FIG. 1, the optical axis O <b> 1 (hereinafter also referred to as the substrate camera optical axis O <b> 1) of the substrate recognition camera 25 and the center line O <b> 3 of the suction nozzle 29 after the component transfer device 26 is replaced. It is necessary to accurately calibrate the positional relationship (distance X4 in the X direction and distance Y4 in the Y direction). In order to accurately correct the misalignment of the component P with respect to the center line O3 of the suction nozzle 29, the suction nozzle center line O3 and the optical axis O2 of the component recognition camera 15 (hereinafter also referred to as the component camera optical axis O2) at the time of detection. The component P needs to be detected by the component recognition camera 15 in a state where the positional relationship with the component recognition is accurately grasped.

  As a method for measuring and calibrating the positional relationship (distance X4 and distance Y4) between the optical axis O1 of the substrate recognition camera 25 and the center line O3 of the suction nozzle 29 of the component transfer device 26, JP-A-7-19816 is disclosed. There is a method described as a prior art in the publication. This is because a component (or jig, the same applies hereinafter) sucked by the suction nozzle is mounted on the substrate, the substrate recognition camera is moved onto the component, and the positional relationship between the optical axis of the substrate recognition camera and the component And calibrates the positional relationship between the optical axis of the substrate recognition camera and the center line of the suction nozzle of the component transfer device based on the measured positional relationship and the amount and direction of movement of the substrate recognition camera. It is. However, this measuring method has a problem that it includes an error in the shape of the part and an error in the suction position to the suction nozzle.

  On the other hand, in the technique described in Japanese Patent Application Laid-Open No. 7-19816, the first and second references that serve as a reference are the positions corresponding to the position where the suction nozzle is attached and the position where the substrate recognition camera is facing. Using the measurement jig provided with the mark, with the measurement jig attached to the suction nozzle so that the first reference mark corresponds, the component recognition camera recognizes the first reference mark, The attached suction nozzle is moved relative to the component recognition camera so that the second reference mark is recognized by the component recognition camera, and then the second reference mark is recognized by the substrate recognition camera. The substrate recognition camera relative to the suction nozzle is determined from the position of the second reference mark recognized in step 1, the positions of the first and second reference marks by the component recognition camera, and the moving distance of the measuring jig And it detects the mounting position. According to this method, it is only necessary to prepare a simple measuring jig, so that it is possible to measure the mounting position of the camera for recognizing the substrate with respect to the suction nozzle with sufficient accuracy, and to reduce the necessary member cost. When the tool is attached to the suction nozzle, the measurement can be automatically performed by an electronic component mounting apparatus or the like, so that there is an advantage that the measurement error due to the degree of shafting of the operator is not included.

Japanese Patent Laid-Open No. 7-19816 (3rd page, FIG. 2)

  However, in the method described in Japanese Patent Laid-Open No. 7-19816, reference marks are provided at positions corresponding to the suction nozzle and the substrate recognition camera, and therefore a measuring jig having a considerable size is used as one reference. Since the suction nozzle is sucked and supported at the position of the mark, the suction nozzle may be deformed due to the gravity of the measurement jig, resulting in a measurement error, or the measurement jig may be detached from the suction nozzle with a slight external force. There has been a problem that it is not always easy to measure the mounting position of the substrate recognition camera with respect to the suction nozzle so that the tool does not come off the suction nozzle.

  The present invention is an electronic component mounting apparatus in which a component transfer device having a rotary head is fixed to a movable table in an exchangeable manner, and each correction value of the rotation center position of each spindle of the rotary head of the replaced component transfer device can be easily obtained. And to find it accurately.

  In order to solve the above-described problem, the structural feature of the invention according to claim 1 is that a substrate transfer device provided on a base for carrying in and out of the substrate and positioning and holding, and an X direction with respect to the base, and Mounting point is a moving table supported so as to be movable in two directions of the Y direction, and a plurality of spindles for picking up components supplied by a component supply device and mounting them on the substrate positioned and supported on the substrate transfer device A component transfer device having a rotary head that can be indexed and rotated and detachably attached to the moving table, a substrate recognition camera fixed to the moving table, and a component recognition camera fixed to the base In this electronic component mounting apparatus, the reference mark provided on the base is imaged by the component recognition camera, and the moving base is stopped at a predetermined position with respect to the coordinate origin to be the same as the reference mark. Is an image of a reference mark having a fixed positional relationship captured by the substrate recognition camera, and the optical axis of the substrate recognition camera measured based on the position of the image of the reference mark captured by the substrate recognition camera Based on the positional relationship between the reference mark and the optical axis of the component recognition camera measured based on the positional relationship with the reference mark and the position of the image of the reference mark captured by the component recognition camera. The coordinate position with respect to the coordinate origin of the optical axis of the component recognition camera is calibrated, and the rotation center of the spindle indexed to the mounting point is positioned at the coordinate position of the optical axis of the calibrated component recognition camera. The carriage is moved to a predetermined position, and each spindle is positioned at a first rotation angle position and a second rotation angle position different from the first rotation angle position. The tip of the suction nozzle mounted on the dollar is imaged by the component recognition camera, and based on the center position of the tip image of each suction nozzle at the first and second rotation angle positions indexed to the mounting point. The rotation center position of each spindle is obtained, and each correction value of the rotation center position of each spindle is obtained from the rotation center position at the mounting point obtained from the image of each spindle and the designed rotation center position.

  In the invention according to claim 1 configured as described above, the reference mark provided on the base is imaged by the component recognition camera. The moving base is stopped at a predetermined position with respect to the coordinate origin, and the reference mark having the same or constant positional relationship as the reference mark is imaged by the substrate recognition camera. The positional relationship between the optical axis of the substrate recognition camera and the reference mark measured based on the position of the image of the reference mark imaged by the substrate recognition camera, and the reference mark imaged by the component recognition camera The coordinate position of the optical axis of the component recognition camera with respect to the coordinate origin is calibrated based on the positional relationship between the optical axis of the component recognition camera measured based on the position of the image and the reference mark. Moves so that the center of rotation of the spindle indexed at the mounting point is positioned at the coordinate position of the calibrated optical axis of the component recognition camera among the multiple spindles held by the rotary head mounted on the moving table. Move the platform to a predetermined position. The tip of the suction nozzle mounted on each spindle is imaged by the component recognition camera with each spindle positioned at a first rotation angle position and a second rotation angle position different from the first rotation angle position. The rotation center position of each spindle is obtained based on the center position of the tip image of each suction nozzle at the first and second rotation angle positions indexed to the mounting point. Each correction value of the rotation center position of each spindle is obtained from the rotation center position at the mounting point obtained from the image of each spindle and the designed rotation center position. Thereby, each correction value of the center position of the plurality of suction nozzles mounted on the rotary head can be obtained efficiently and accurately. Further, the fluttering of the suction nozzle can be corrected.

The figure which shows the positional relationship of the principal part of the electronic component mounting apparatus which can replace | exchange several types of different component transfer apparatuses. The partial expansion perspective view explaining the principal part of the method of calibrating the positional relationship between the optical axis of the substrate recognition camera and the center line of the suction nozzle. The figure explaining the method of calibrating the positional relationship shown in FIG. The partial expansion perspective view explaining the principal part of the other method which calibrates the positional relationship of the optical axis of the camera for board | substrate recognition, and the centerline of a suction nozzle. The perspective view which shows the principal part of the electronic component mounting apparatus which can replace | exchange several types of different component transfer apparatuses. The sectional side view which shows the support structure of the upper end member which supports a reference gauge, and a component recognition camera. The figure which shows the components transfer apparatus provided with the rotary head which concerns on embodiment of this invention. The figure which shows the image of the end surface of a some suction nozzle. The figure which shows the image of the components adsorbed | sucked by the some suction nozzle. The figure which shows the amount of displacement of the components center position by a suction error when rotating a suction nozzle. The figure explaining the principal part of the calibration method which concerns on embodiment of this invention, and is the figure which shows the image imaged by positioning the suction nozzle indexed to the mounting point on the component camera optical axis O2. The figure which shows the descent | fall error which is a difference of the coordinate position of the front-end | tip center of a suction nozzle in the raising position and descending position of each spindle of a rotary head.

  A method for calibrating the positional relationship between the optical axis of the substrate recognition camera and the center line of the suction nozzle in the electronic component mounting apparatus shown in FIGS. 1 to 6 will be described below. A plurality of electronic component mounting apparatuses 10 having a schematic overall structure shown in FIG. 5 to which this method is applied are arranged side by side to constitute a component mounting line. On the base 11 of each electronic component mounting apparatus 10, two board transfer apparatuses 12 and 12 a that transfer the boards S and Sa in the Y direction are provided. Although not shown in the drawings, the electronic component mounting apparatuses 10 are arranged adjacent to each other so that the respective substrate transfer apparatuses 12 and 12a are continuous in the Y direction. 12a is operated in conjunction with each other so that the substrates S and Sa are sequentially fed onto the adjacent substrate transfer devices 12 and 12a, and are positioned and held at predetermined positions.

  On the upper side of each substrate transfer device 12, 12 a, a slide 21 that is elongated in the Y direction is fixed to the lower surface of a table 30 that is movably guided and supported by a fixed rail 20 that extends in the X direction perpendicular to the Y direction. The movement in the X direction is controlled by a servo motor 22 via a ball screw. On one side of the slide 21, a moving table 24 to which a substrate recognition camera 25 and a component transfer device 26 are attached is guided and supported so as to be movable in the Y direction, and the movement is controlled by a servo motor 31 via a ball screw. ing. The slide 21, the moving base 24, the substrate recognition camera 25 and the component transfer device 26 attached to the slide 21, and the component transfer device 26 are provided on the upper sides of both the substrate transfer devices 12 and 12a, respectively. Therefore, in FIG. 5, the upper part of the substrate transfer device 12a is not shown. In the following description, the structure and operation of the slide 21 and the moving table 24 will be described only on the substrate transfer apparatus 12 side. Note that this method of calibrating this positional relationship is not limited to the electronic component mounting apparatus 10 provided with two board transfer apparatuses, but can be applied to the electronic component mounting apparatus 10 provided with one board transfer apparatus.

  The component transfer device 26 attached to the moving table 24 has a plurality of types of different performances for mounting high-speed small-sized electronic components and irregular-shaped large-sized electronic components, and can be exchanged for the moving table 24. The electronic component mounting apparatus 10 is composed of two parts: a component transfer apparatus 26 that can be exchanged in this way, and a mounting apparatus basic part that includes all other parts.

  As shown in FIGS. 1, 2, and 5, each component transfer device 26 includes a support base 27 that is detachably attached to the movable table 24, and a Z direction perpendicular to the X and Y directions on the support base 27. A component mounting head 28 that is guided and supported by a servo motor 32 through a ball screw and is controlled by a servo motor 32, and a cylinder that protrudes downward from the component mounting head 28 and holds the component P at the lower end by suction. The suction nozzle (part collecting unit) 29 is shaped like a tube. The center line O3 of the component mounting head 28 and the suction nozzle 29 is parallel to the Z direction, and the suction nozzle 29 is supported so as to be rotatable around the center line O3 with respect to the component mounting head 28 and is driven by a servo motor (not shown). The rotation angle is controlled. The substrate recognition camera 25 is fixed to the moving base 24 and is not replaced except in the case of a failure, and its optical axis O1 is parallel to the Z direction.

  On one end side of the electronic component mounting apparatus 10, a component supply apparatus 13 including a plurality of feeders installed side by side is provided. A component recognition camera 15 having an optical axis O2 parallel to the Z direction is provided on the base 11 between the substrate transfer device 12 and the component supply device 13. As shown in FIG. 6, the component recognition camera 15 is mounted on a base 11 via a support base 16, and a bowl-shaped upper end member (support member) 17 having an open top and no bottom is provided above the camera 15. It is coaxially attached to the optical axis O2 via the connecting portion 16a. The open upper surface of the upper end member 17 is covered with a transparent cover glass 18, and the side light source 19 made up of a number of LEDs is provided on the inner surface of the upper end member 17, and the component P and the component P recognized by the component recognition camera 15 are provided. The lower end of the suction nozzle 29 is illuminated from below.

  In FIG. 1 and FIG. 5, the electronic component mounting apparatus 10 detects the position of the board mark Sm provided on the board S carried in and held by the board carrying apparatus 12 by the board recognition camera 25, and this board. The position correction is performed based on the position of the mark Sm, the slide 21 and the moving table 24 are moved in the X direction and the Y direction, and the electrons held by suction from the component supply device 13 to the tip of the suction nozzle 29 of the component transfer device 26. The component P is mounted on the commanded coordinate position on the substrate S. The suction nozzle 29 is temporarily stopped on the component recognition camera 15 while the component P held by the tip of the suction nozzle 29 is moved from the component supply device 13 to the commanded coordinate position on the substrate S. The component recognition camera 15 detects the misalignment of the component P with respect to the center line O3 of the suction nozzle 29 and the angle shift around the center line O3. Then, after the suction nozzle 29 is rotated by the servo motor based on the detection result of the angular deviation to correct the angular deviation around the center line O3, the X direction of the slide 21 and the movable table 24 and the moving table 24 are determined based on the misalignment detection result. The amount of movement in the Y direction is corrected, and the component P is accurately mounted at the commanded coordinate position (X1, Y1 in FIG. 1) based on the board mark Sm on the board S.

  As described above, the electronic component mounting apparatus 10 is composed of the mounting apparatus basic portion and the component transfer apparatus 26 that is replaceably attached to the moving table 24. Therefore, the electronic component mounting apparatus 10 is mounted on the substrate S with the substrate mark Sm as a reference. In order to accurately mount the component P at the predetermined coordinate position (X1, Y1), the optical axis O1 of the substrate recognition camera 25 and the suction nozzle of the component transfer device 26 after the component transfer device 26 is replaced. It is necessary to accurately calibrate the positional relationship between the 29 center lines O3 (distance X4 in the X direction and distance Y4 in the Y direction). Further, in order to accurately detect the misalignment of the component P with respect to the center line O3 of the suction nozzle 29, the positional relationship between the suction nozzle center line O3 and the optical axis O2 of the component recognition camera 15 is accurately grasped at the time of detection. It is necessary to detect the component P by the component recognition camera 15. For this purpose, it is necessary to accurately calibrate the positional relationship (distance X2 in the X direction and distance Y2 in the Y direction) between the coordinate origin of the electronic component mounting apparatus 10 and the component camera optical axis O2. Further, it is necessary to calibrate when the component recognition camera 15 is replaced.

  Next, the operation of the method for calibrating such a positional relationship will be described mainly with reference to FIGS. Prior to the start of the operation, as shown in FIGS. 1 and 3, a reference gauge G provided with various reference marks Gm such as a circle and a cross on a colorless and transparent glass plate is attached to a cover glass of the upper end member 17. The reference mark Gm is placed on the part 18 so as to fall within the field of view of the component recognition camera 15. The movable table 24 is positioned at a predetermined position so that the reference mark Gm is within the field of view of the substrate recognition camera 25 and the tip of the suction nozzle 29 of the component transfer device 26 is within the field of view of the component recognition camera 15. Is stopped. The positional relationship (distance X3 and distance Y3 in FIG. 3) of the board camera optical axis O1 with respect to the coordinate origin of the electronic component mounting apparatus 10 at that time is the movement of the slide 21 and the moving base 24 of the electronic component mounting apparatus 10 from the coordinate origin. The quantity is detected by a position detection device such as induct thin and recorded in the memory of the control device 23. In this stopped state, the reference mark Gm is within the depth of focus of the component recognition and board recognition cameras 15 and 25, and the tip of the suction nozzle 29 also approaches the reference gauge G and the component recognition camera 15 Is moved down to a position (see a two-dot chain line 29a in FIG. 2) that falls within the depth of focus.

  In this way, when the moving base 24 is positioned at a predetermined position so that the substrate camera optical axis O1 moves from the coordinate origin to the distance X3 and the distance Y3 in FIG. 3, the substrate recognition camera 25 is positioned at the position of the image of the reference mark Gm. , The positional relationship (distance Xa and distance Ya) between the substrate camera optical axis O1 and the reference mark Gm is measured. The component recognition camera 15 measures the positional relationship (distance Xc and distance Yc) between the component camera optical axis O2 and the suction nozzle center line O3 based on the position of the image of the tip of the suction nozzle 29 and the reference mark Gm. Based on the position of the image, the positional relationship (distance Xb and distance Yb) between the component camera optical axis O2 and the reference mark Gm is measured.

As described above, the light of the substrate recognition camera 25 after replacing the component transfer device 26 necessary for accurately mounting the component P at a predetermined coordinate position on the substrate S with respect to the substrate mark Sm. As is apparent from FIGS. 2 and 3, the calibration value of the positional relationship (distance X4 in the X direction and distance Y4 in the Y direction) between the axis O1 and the center line O3 of the suction nozzle 29 of the component transfer device 26 is as follows. Formula X4 = Xa + Xb + Xc (1a)
Y4 = Ya + Yb + Yc (1b)
Given by.

When the moving base 24 is positioned at a predetermined position so that the substrate camera optical axis O1 moves from the coordinate origin by the distance X3 and the distance Y3 in FIG. 3, the optical axis O1 of the substrate recognition camera 25 and the optical axis of the component recognition camera 15 As is clear from FIG. 3, the calibration value of the positional relationship with O2 (distance X5 in the X direction and distance Y5 in the Y direction) is expressed by the following formula: X5 = Xa + Xb (2a)
Y5 = Ya + Yb (2b)
Given by.

When the moving base 24 is positioned at a predetermined position so that the substrate camera optical axis O1 moves from the coordinate origin by the distance X3 and the distance Y3 in FIG. 3, the positional relationship of the component camera optical axis O2 with respect to the coordinate origin (distance in the X direction). As is apparent from FIG. 3, the calibration values of the distance Y2) in the X2 and Y directions are X5 and Y5 given by the above equation and the substrate camera light with respect to the coordinate origin recorded in the memory of the control device as described above. Based on the positional relationship X3, Y3 of the axis O1, the following expression X2 = X3 + X5 (3a)
Y2 = Y3 + Y5 (3b)
Given by.

  Thus, when the component P sucked by the suction nozzle 29 is imaged by the component recognition camera 15, the position of the center line of the component P from the coordinate origin is determined from the positional relationship of the component camera optical axis O2 with respect to the coordinate origin. Finally, the deviation of the sucked part P with respect to the suction nozzle center O3 can be calculated by the control device 23.

  Note that the positive and negative signs such as the distances Xa, Xb,... In the above equations vary depending on the positional relationship between the center line O3 and the optical axes O1, O2.

  After the replacement of the component transfer device 26, the control device 23 that controls the operation of the electronic component mounting device 10 first moves the moving base 24 to the predetermined position shown in FIGS. With the reference gauge G placed on the upper end member 17 of the camera 15, the distance Xa, Ya is measured by the board recognition camera 25, and the distance Xb, Yb and the distance Xc, Yc are measured by the component recognition camera 15. . Next, the control device calculates the calibration values of the distances X4 and Y4 by the equations (1a) and (1b), calculates the calibration values of the distances X5 and Y5 by the equations (2a) and (2b), and the equations (3a), ( The calibration values of the distances X2 and Y2 are calculated according to 3b).

  Then, when the substrate S is carried in by the substrate transport device 12, the control device 23 moves the slide 21 and the moving table 24 in the X direction and the Y direction, and removes a predetermined component P from the component supply device 13 to the suction nozzle 29. Adsorb and hold at the tip. Next, the moving table 24 is moved to a predetermined position, and the component P sucked by the suction nozzle 29 is imaged by the component recognition camera 15. The distance from the coordinate origin of the center line of the image of the component P is obtained based on calibration values X2 and Y2 of the positional relationship of the component camera optical axis O2 with respect to the coordinate origin. The distance from the coordinate origin of the suction nozzle center line O3 obtained from the calibration values X4 and Y4 of the positional relationship between the substrate camera optical axis O1 and the suction nozzle center line O3 and the distance from the coordinate origin of the center line of the component P The difference is calculated, the suction displacement of the sucked component P with respect to the suction nozzle center O3 is calculated, the suction displacement is corrected, and the component P is accurately mounted at the command position on the substrate. At this time, correction due to the positioning error of the substrate S calculated based on the deviation of the substrate mark Sm imaged by the substrate recognition camera 25 is also performed at the same time. When the component recognition camera 15 detects an angle shift around the center line O3 of the component P, the suction nozzle 29 is rotated to correct the angle shift of the component P.

  In this case, a plurality of reference marks Gm, for example, four reference marks Gm are provided on the reference gauge G so as to be simultaneously spaced within the field of view of the component recognition camera 15 and are appropriately separated in the X and Y axis directions. Thus, if each calibration value is calculated | required and an average value is calculated, a more accurate calibration value can be obtained.

  In the above-described method of calibrating the positional relationship, the reference mark Gm of the reference gauge G placed on the upper end member 17 is within the field of view of the cameras 15 and 25 for component recognition and board recognition. When the center line O3 of 29 and the optical axis O1 of the substrate recognition camera 25 are separated to some extent, the reference mark Gm and the tip of the suction nozzle 29 are placed in the field of view of the component recognition camera 15 and at the same time the substrate recognition camera. The reference mark Gm cannot be put in the 25 field of view. In that case, the reference gauge G is placed on the upper end member 17 so that the reference mark Gm falls within the field of view of the component recognition camera 15, and the moving table 24 so that the reference mark Gm falls within the field of view of the board recognition camera 25. Is stopped at the first position. The board camera optical axis O1 and the component camera optical axis O2 at the first position are determined from the positional relationship between the optical axes O1 and O2 of each camera and the reference mark Gm by the component recognition camera 15 and the board recognition camera 25 at the first position. The positional relationship X5, Y5 is detected. Next, the moving base 24 is moved to the second position so that the tip of the suction nozzle 29 falls within the field of view of the component recognition camera 15, and the component recognition camera 15 causes the component camera optical axis O2 and the suction nozzle center line O3 to be The positional relationship Xc, Yc is detected, the positional relationship X5, Y5 between the board camera optical axis O1 and the component camera optical axis O2 detected at the first position, and the component camera optical axis O2 detected at the second position. The positional relationship X4 between the substrate camera optical axis O1 and the suction nozzle central line O3 based on the positional relationship Xc, Yc with the nozzle center line O3 and the positional relationship Xe, Ye (not shown) between the first position and the second position. , Y4 may be calculated. In this case, the values of the distances X4 and X5 are the values calculated by the equations 1a and 2a plus Xe, and the values of the distances Y4 and Y5 are the values calculated by the equations 1b and 2b plus Ye. However, the other points are as described above.

  In the modified example of the method of calibrating such a positional relationship, each distance Xa, Ya, Xb, Yb, Xc, Yc must be measured twice at the first position and the second position of the moving table 24. It takes time and effort. However, the second method for calibrating the positional relationship between the optical axis of the substrate recognition camera and the center line of the suction nozzle shown in FIG. 4 has a certain distance between the suction nozzle center line O3 and the substrate camera optical axis O1. Even in this case, the positional relationship between the substrate camera optical axis O1 and the component camera optical axis O2 is calibrated and the substrate camera light is maintained while the movable table 24 is stopped at one predetermined position as in the case described with reference to FIGS. Calibration of the positional relationship between the axis O1 and the suction nozzle center line O3 can be performed.

  The second method for calibrating this positional relationship is that, instead of the above-described reference gauge G, a reference gauge Ga provided with first and second reference marks Gm1, Gm2 at a distance L is used. This is different from the case described in FIG. Next, the operation of the second method will be described. Also in the second method, prior to the calibration operation, for example, the first and second reference marks are arranged so that the first and second reference marks Gm1 and Gm2 have a predetermined positional relationship as shown in FIG. The reference gauge Ga is placed on the cover glass 18 of the upper end member 17 of the upper end member 17 so that the straight line connecting Gm1 and Gm2 is parallel to the Y axis and the first reference mark Gm1 is within the field of view of the component recognition camera 15. Placed. The moving table 24 is stopped at a predetermined position so that the second reference mark Gm2 is in the field of view of the board recognition camera 25 and the tip of the suction nozzle 29 is in the field of view of the component recognition camera 15. The positional relationship of the optical axis O1 of the substrate recognition camera 25 with respect to the coordinate origin at that time (same as the distance X3 and the distance Y3 in FIG. 3) is recorded in the memory of the control device 23 that controls the operation of the electronic component mounting apparatus 10. Yes.

  In this state, the component recognition camera 15 has a positional relationship (distance Xc and distance Yc) between the optical axis O2 and the suction nozzle center line O3, and a positional relationship (distance Xb) between the optical axis O2 and the first reference mark Gm1. And distance Yb). The substrate recognition camera 25 measures the positional relationship (distance Xa and distance Ya) between the optical axis O1 and the second reference mark Gm2.

In the second method of calibrating this positional relationship, the positional relationship between the optical axis O1 of the substrate recognition camera 25 after the replacement of the component transfer device 26 and the center line O3 of the suction nozzle 29 of the component transfer device 26 ( As is clear from FIG. 4, the calibration values of the distance X4 in the X direction and the distance Y4) in the Y direction are expressed by the following equation: X4 = Xa + Xb + Xc (1a)
Y4 = Ya + Yb + Yc + L (1b ′)
And the positional relationship between the optical axis O2 of the component recognition camera 15 and the optical axis O1 of the board recognition camera 25 when the moving body 24 is located at a predetermined position (distance X5 in the X direction and distance in the Y direction). The calibration value of Y5) is the following formula: X5 = Xa + Xb (2a)
Y5 = Ya + Yb + L (2b ′)
Given by.

The calibration value of the positional relationship of the component camera optical axis O2 with respect to the coordinate origin (distance X2 in the X direction and distance Y2 in the Y direction) is stored in the memory of the control device 23 as described above with X5 and Y5. Based on the positional relationship X3, Y3 of the substrate camera optical axis O1 with respect to the coordinate origin when the recorded moving base 24 is located at a predetermined position, the following expression X2 = X3 + X5 (3a)
Y2 = Y3 + Y5 (3b)
Given by. As described above, the positive and negative signs such as the distances Xa, Xb,... In the above equations are different depending on the positional relationship between the center line O3 and the optical axes O1, O2.

  After the replacement of the component transfer device 26, the control device 23 that controls the operation of the electronic component mounting device 10 first moves the moving base 24 to the predetermined position shown in FIG. With the reference gauge Ga placed on the upper end member 17 in a predetermined positional relationship, the distances Xa and Ya are measured by the board recognition camera 25, and the distances Xb and Yb and the distances Xc and Yc are measured by the component recognition camera 15. taking measurement. Next, the control device 23 calculates the calibration values of the distances X4 and Y4 by the equations (1a) and (1b ′), calculates the calibration values of the distances X5 and Y5 by the equations (2a) and (2b ′), and the equation (3a ) And (3b), the calibration values of the distances X2 and Y2 are calculated. The mounting of the component P on the substrate S is the same as that described with reference to FIGS.

  In this second method, the reference gauge Ga is placed on the upper end member 17 so that the first and second reference marks Gm1, Gm2 are completely aligned in the Y direction. The reference gauge Ga may be placed on the upper end member 17 so that the directions of the marks Gm1 and Gm2 are inclined at a predetermined angle with respect to the Y direction (or X direction). In this case, X4, X5 and Y4, Y5 of the second method are the X-direction component of the distance L and the Y-direction with respect to X4, X5 and Y4, Y5 in the method of calibrating the positional relationship shown in FIGS. A value obtained by adding only the components may be used.

  In the method of calibrating each positional relationship, the positional relationship X4, Y4 between the optical axis O1 of the substrate recognition camera 25 and the center line O3 of the suction nozzle 29, and the optical axis of the component recognition camera 15 with respect to the coordinate origin are used. Although it has been described that the reference gauges G and Ga are placed on the upper end member 17 each time the positional relationship X2 and Y2 of O2 is calibrated, the reference gauges G and Ga may always be placed on the upper end member 17. Good. Alternatively, one or two reference marks may be provided on the cover glass 18 of the upper end member 17 and used as the reference gauges G and Ga.

  Next, an embodiment of the present invention will be described in which the component transfer device 40 having a rotary head is detachably attached to the moving table 24. As shown in FIG. 7, the component transfer device 40 includes a head frame 41 attached to the moving table 24. A cylindrical rotary head 42 that holds a plurality of (for example, eight) spindles 44n so as to reciprocate in the vertical direction (Z-axis direction) is attached to the lower portion of the head frame 41 so as to be rotatable about an axis. Each spindle 44n is urged upward by a compression spring (not shown), and a suction nozzle 29n is attached to the lower end on the same axis.

  The rotary head 42 is intermittently rotated by a servo motor 43 attached to the head frame 41 so that each suction nozzle 29n stops at a predetermined position. The spindle 44n stopped at the mounting point (mounting station) where the electronic component P is mounted in a predetermined position, when the nozzle lowering lever 47 is lowered by the rotation of the feed screw 46 driven by the servo motor 45, the spring of the compression spring. The suction nozzle 29n is lowered against the force. When the nozzle lowering lever 47 is raised by the rotation of the feed screw 46 in the opposite direction, the spindle 44n is raised by the spring force of the compression spring, and the suction nozzle 29n is also raised. Further, all the spindles 44n are rotationally connected to the servo motors 48, and the respective spindles 44n and the suction nozzles 29n are simultaneously rotated around the respective axes by the servo motors 48. When the component Pn sucked by the suction nozzle 29n of the spindle 44n indexed to the mounting point is rotated by α degrees with respect to the X axis and mounted on the substrate S, the spindle 44n is moved by the servo motor 48. It is rotated α degrees. Each suction nozzle 29n is connected to a negative pressure supply source via a pipe line provided with an on-off valve (both not shown). The suction nozzle 29n is selected and used depending on the type of electronic component Pn to be sucked.

  Next, the case where the tips of the plurality of nozzles 29n are simultaneously imaged by the component recognition camera 15 and the rotation center position OR of the rotary head 42 is calibrated will be described. Since the rotational center OR of the rotary head 42 and the optical axis O1 of the board recognition camera 25 are considerably separated, the reference mark Gm and the tips of all the suction nozzles 29n of the rotary head 42 are placed in the field of view of the component recognition camera 15. At the same time, the reference mark Gm cannot be placed in the field of view of the substrate recognition camera 25. Accordingly, the reference gauge G is placed on the upper end member 17 so that the reference mark Gm is within the field of view of the component recognition camera 15, and the movable table 24 is placed so that the reference mark Gm is within the field of view of the board recognition camera 25. Stop at 1 position. At this time, the board camera optical axis O1 is located at the coordinate position (X3, Y3) with respect to the coordinate origin of the electronic component mounting apparatus 10. The board camera optical axis O1 and the component camera optical axis O2 at the first position are determined from the positional relationship between the optical axes O1 and O2 of each camera and the reference mark Gm by the component recognition camera 15 and the board recognition camera 25 at the first position. Are calculated by the equations (2a) and (2b), and the coordinate position (X2, Y2) with respect to the coordinate origin of the component camera optical axis O2 is obtained from the equations (3a) and (3b). Next, the moving table 24 is moved to the second position by the servo motors 22 and 31 so that the rotation center OR of the rotary head 42 is located at the coordinate position (X2, Y2). At this time, a value stored in the control device 23 as a design value is used as the distance between the optical axis O1 of the substrate recognition camera and the rotary head rotation center OR.

  With the rotary head 42 indexed at the imaging position, the tips of all the suction nozzles 29n are imaged by the component recognition camera 15 whose optical axis O2 is located at the coordinate position (X2, Y2) (see FIG. 8). From the image, the coordinate position (Xn1, Yn1) of the tip center of each suction nozzle 29n when each spindle 44n is at the first rotation angle position is obtained. Since each suction nozzle 29n is slightly bent, the rotation center position of each spindle 44n, which is the rotation center position of each suction nozzle 29n, does not necessarily match the tip center position of each suction nozzle 29n at the first rotation angle position. In order to obtain the rotation center position of each spindle 44n, which is the rotation center position of each suction nozzle 29n, first, each spindle 44n is positioned at the first rotation angle position as described above, and the coordinates of the tip center of each suction nozzle 29 are set. The position (Xn1, Yn1) is obtained, then each spindle 44n is rotated 180 degrees by the servo motor 48 and positioned at the second rotation angle position, and the second of each suction nozzle 29n is obtained from the image captured by the component recognition camera 15. The coordinate position (Xn2, Yn2) of the tip center at the rotation angle position is obtained. The coordinate position obtained by arithmetically averaging the coordinate positions (Xn1, Yn1) and (Xn2, Yn2) of the tip center of each suction nozzle 29n at the first and second rotation angle positions is the rotation center position of each suction nozzle 29n. It is obtained as the rotation center position (Xn, Yn). In this case, if the interval between the coordinate positions (Xn1, Yn1) and (Xn2, Yn2) of the tip center of each suction nozzle 29n at the first and second rotation angle positions is greater than or equal to the threshold value, the flare of the suction nozzle 29n is greater than the allowable value. The replacement is instructed by an alarm display or the like. In addition, when the difference in the coordinate position of the tip center of each suction nozzle 29 at the first and second rotation angle positions is small, the coordinate position of the tip center of each suction nozzle 29 obtained from the image captured at the first angle position is set. The coordinate position of each spindle 44n may be used.

  The coordinate position (Xor, Yor) of the rotation center OR of the rotary head 42 is obtained from the coordinate position of the rotation center OR of the spindle 44n, which is the rotation center of each suction nozzle 29n, and the coordinate position (Xor of the rotation center OR of the rotary head 42 is obtained. , Yor) and the coordinate position (X2, Y2) of the optical axis O2 of the component recognition camera 15, the correction value of the positional deviation of the rotational center OR of the rotary head 42 is obtained from the equations ΔXor = Xor−X2 and ΔYor = Yor−Y2. Desired. For the coordinate position (Xor, Yor) of the rotation center OR of the rotary head 42, for example, four sets of intermediate coordinates of the rotation center positions (Xn, Yn) of the two spindles 44n facing each other across the rotation center OR are obtained. It can be obtained from the average value of the intermediate coordinates.

  Then, when the substrate S is carried in by the substrate transport device 12, the control device 23 moves the slide 21 and the moving base 24 in the X direction and the Y direction, rotates the rotary head 42, and mounts each component on the substrate S. Pn is sucked and held from the component supply device 13 to the tip of each suction nozzle 29n. Next, the moving base 24 is moved to the second position, the rotary head 42 is indexed to the imaging position, and the component Pn sucked by each suction nozzle 29n is imaged by the component recognition camera 15 (see FIG. 9). The coordinate position (Xpn, Ypn) from the coordinate origin of the mounting center position of each component Pn that should coincide with the rotation center of the spindle 44n when each component Pn is attracted to each suction nozzle 29n is on the image of each component Pn. Is obtained on the basis of the center position and the coordinate position (X2, Y2) of the component camera optical axis O2. Based on the mounting center position (Xpn, Ypn) of each component Pn and the rotation center position (Xn, Yn) of each spindle 44n, the suction error of each component Pn with respect to the rotation center of each spindle 44n is expressed by the equation ΔXpn = Xpn−Xn. , ΔYpn = Ypn−Yn. The adsorption errors ΔXpn and ΔYpn of each component Pn are decomposed into a radial component ΔRpn connecting the rotation center of the rotary head 42 and the rotation center of each spindle 44n, and a component ΔTpn in a direction perpendicular to the radius, and each spindle 44n serves as a mounting point. A radial component ΔRpn is used as a correction value for the suction error in the X-axis direction when indexed, and a component ΔTpn in a direction perpendicular to the radius is used as a correction value for the suction error in the Y-axis direction. Each spindle 44n is indexed to a mounting point by the rotation of the rotary head 42, and the moving base 24 is positioned by correction amounts ΔXor and ΔYor of the rotation center OR of the rotary head 42 and correction values ΔRpn and ΔTpn of the suction error of each component Pn. The corrected component is moved to the mounting position, and the component Pn is mounted on the substrate S.

Further, when the component P is rotated by α degrees with respect to the X axis and mounted on the substrate S, the component center is displaced by suction errors ΔRpn and ΔTpn which are deviations of the component center with respect to the rotation center of the spindle 44n. When rotating 44 by α degrees, the position of the moving base 24 is corrected by the amount of displacement dXαn, dYαn and mounted. The displacement amounts dXαn and dYαn are as shown in FIG.
dXαn = ΔRpn−Lcos (θ + α), dYαn = ΔTpn−Lsin (θ + α),
L ² = ΔRpn ² + ΔTpn ² , tanθ = ΔTpn / ΔRpn
It can be calculated from

  In the component transfer apparatus 40 including the rotary head 42, as described above, the slide 21 and the moving base 24 are moved to the second position, the rotary head 42 is indexed to the imaging position, and all the parts are raised to the raised position. The tip of the suction nozzle 29n of the spindle 44n is imaged by the component recognition camera 15, and the rotation center position (Xn, Yn) of each spindle 44n, which is the rotation center position of each suction nozzle 29n, is obtained in advance and stored in the control device 23. The Further, the component transfer device 40 is obtained from the difference between the rotation center OR (Xor, Yor) of the rotary head 42 and the coordinate position (X2, Y2) of the component camera optical axis O2 when the movable table 24 is moved to the second position. Mounting error is required.

  When the component P is mounted on the substrate S, the spindle 44n is indexed to the mounting point by the rotation of the rotary head 42, and the slide 21 and the moving base 24 are moved to attract the spindle 44n indexed to the mounting point. Positioned at the storage position of the component supply device 13 in which the component Pn to be suctioned by the nozzle 29n is stored, the spindle 44n is lowered, and the component Pn is suctioned by the suction nozzle 29n and is raised. When all the spindles 44n are lifted by sucking the parts Pn to the respective suction nozzles 29n, the slide 21 and the moving base 24 are moved to the second position, the rotary head 42 is indexed to the imaging position, and all the rising spindles 44n The component Pn sucked by the suction nozzle 29n is imaged by the component recognition camera 15.

  The coordinate position (Xpn, Ypn) of the mounting center position of each component Pn is obtained from the image of each component Pn, the coordinate position (Xpn, Ypn) of the mounting center position of each component Pn and the rotation center position (Xn) of each spindle 44n. , Yn), a suction error with respect to the rotation center of each spindle 44n of each component Pn is obtained. In order to accurately mount the component Pn sucked at the tip of the suction nozzle 29n of the spindle 44n, which is indexed to the mounting point, at the command position on the substrate S, the slide 21 and the moving table 24 are attached with an error in mounting the component transfer device 40. Then, the position is corrected by the suction error of the component Pn and moved to the mounting position, the spindle 44n is lowered to the lowered position, and the component Pn is mounted on the substrate S.

  The spindle 44n is guided by the rotary head 42 so as to be able to reciprocate in the vertical direction and is moved between the raised position and the lowered position, so that the tip of the suction nozzle 29n guides the spindle 44n at the raised position and the lowered position of the spindle 44n. Displace with accuracy. The amount of displacement of the tip of the suction nozzle 29n at the ascending position and the descending position of the spindle 44n is obtained as a descending error of each spindle 44n, and the slide 21 and the moving table 24 are attached to the component transfer device 40 and the component Pn. If the position is corrected by a correction value obtained by adding a lowering error to the mounting position and moved to the mounting position, the suction nozzle 29n can mount the component Pn more accurately at the command position on the substrate S in the lowering position.

  In order to measure the descent error, the slide 21 and the moving table 24 are moved to the second position, and the tips of the suction nozzles 29n of all the spindles 44n rising to the ascending position are imaged by the component recognition camera 15, and each spindle When the rotation center position (Xn, Yn) of 44n is obtained in advance, as shown in FIG. 12, the tip of the suction nozzle 29n attached to the spindle 44n indexed to the mounting point is imaged by the component recognition camera 15. Then, the coordinate position (Xnup, Ynup) of the tip center of the suction nozzle 29n at the ascending position is obtained, then the spindle 44n is lowered to the lowered position, and the tip of the suction nozzle 29n is imaged by the component recognition camera 15 and lowered. The coordinate position (Xndn, Yndn) of the tip center of the suction nozzle 29n is obtained, and the descending errors are ΔXnd = Xndn−Xnup, ΔYnd = Yndn It is calculated from Ynup. In this way, the lowering error of the suction nozzle 29n for all the spindles 44n is obtained in advance and stored in the control device 23. The rotation center position of the suction nozzle 29n is obtained by arithmetically averaging the coordinate positions of the tip centers of the suction nozzles 29n at the first and second rotation angle positions, but the descent error is determined by each spindle 29n at one rotation angle position. It can be obtained from the difference in the coordinate position of the tip center of the suction nozzle 29n between the ascending position and the descending position.

  Next, the plurality of nozzles 29n of the rotary head 42 are sequentially indexed to the mounting point, and each tip is individually imaged by the component recognition camera 15, and the rotation center position of the spindle 44n, which is the rotation center position of each suction nozzle 29n, is determined. The case of calibrating will be described. Also in this case, the rotation center OR of the rotary head 42 and the optical axis O1 of the board recognition camera 25 are considerably separated from each other, so that the suction marks indexed to the reference mark Gm and the mounting point are within the field of view of the component recognition camera 15. At the same time as inserting the tip of the nozzle 29n, the reference mark Gm cannot be placed in the field of view of the substrate recognition camera 25. Accordingly, similarly to the above-described case, the moving base 24 is stopped at the first position, the reference mark Gm is imaged by the component recognition camera 15 and the board recognition camera 25, and the board camera optical axis O1 at the first position is obtained. The positional relationship X5, Y5 with the component camera optical axis O2 is calculated by the equations (2a), (2b), and the coordinate position (X2, Y2) with respect to the coordinate origin of the component camera optical axis O2 is calculated by the equations (3a), (3b). Ask from. Next, the moving table 24 is moved to the second position by the servo motors 22 and 31 so that the rotation center of the spindle 44 indexed to the mounting point of the rotary head 42 is located at the coordinate position (X2, Y2). At this time, a value stored in the control device 23 as a design value is used as the distance between the optical axis O1 of the board recognition camera and the rotation center of the spindle 44 determined at the mounting point.

  The tip of each suction nozzle 29n that is sequentially indexed to the mounting point is picked up by the component recognition camera 15 whose optical axis O2 is located at the coordinate position (X2, Y2) (see FIG. 11), and each spindle 44n from the image. The coordinate position (Xn1, Yn1) of the tip center of each suction nozzle 29n at the first rotation angle position is obtained. Since each suction nozzle 29n is slightly bent, the rotation center position of each spindle 44n does not necessarily match the tip center position of each suction nozzle 29n at the first rotation angle position. In order to obtain the correction value of the rotation center position of each spindle 44n, each spindle 44n is sequentially indexed to the mounting point, and as described above, each spindle 44n is positioned at the first rotation angle position and the tip center of each suction nozzle 29n is positioned. Coordinate position (Xn1, Yn1) is obtained, each spindle 44n is rotated 180 degrees by servo motor 45 and positioned at the second rotation angle position, and each spindle 44n is sequentially indexed to the mounting point to obtain the second rotation angle. The coordinate position (Xn2, Yn2) of the tip center of each suction nozzle 29n at the position is obtained. The coordinate positions (Xn1, Yn1) and (Xn2, Yn2) of the tip center of each suction nozzle 29n at the first and second rotation angle positions are arithmetically averaged to obtain the rotation center position (Xn, Yn) of each spindle 44n sequentially. . As the correction values ΔXn and ΔYn of the rotation center position of each spindle 44n, the rotation center position (Xn, Yn) of each spindle 44n obtained from the coordinate position of the tip center of the nozzle 29n and the coordinate position of the component camera optical axis O2 ( X2, Y2) and the difference ΔXn = Xn−X2, ΔYn = Yn−Y2 are obtained. In this case, when the distance between the coordinate positions (Xn1, Yn1) and (Xn2, Yn2) of the tip center of each suction nozzle 29n at the first and second rotation angle positions is greater than or equal to the threshold value, the touch of the suction nozzle 29 is greater than the allowable value. The replacement is instructed by an alarm display or the like.

  Then, when the substrate S is carried in by the substrate transport device 12, the control device 23 moves the slide 21 and the moving table 24 in the X direction and the Y direction, and each component Pn to be mounted on the substrate S from the component supply device 13. It sucks and holds at the tip of each suction nozzle 29n. Next, the moving base 24 is moved to the second position, each spindle 44n is sequentially indexed to the mounting point, and each component Pn sucked by each suction nozzle 29n is imaged by the component recognition camera 15. The coordinate position (Xpn, Ypn) of the center point of the image of each component Pn is obtained, and the difference ΔXpn = Xpn−Xn, ΔYpn = Ypn−Yn from the coordinate position (Xn, Yn) of the rotation center of each spindle 44n It is obtained as suction errors ΔXpn and ΔYpn with respect to the rotation center of each suction nozzle 29n of the part Pn. The moving base 24 moves to each mounting position by correcting the position by the correction values ΔXn and ΔYn of the rotation center position of each spindle 44n and the correction values ΔXpn and ΔYpn of the suction error of each component Pn, which are indexed to the mounting point. Then, each component Pn is sequentially mounted on the substrate S.

When the component P is rotated by α degrees with respect to the X axis and mounted on the substrate S, the component center is displaced by the suction error ΔXpn, ΔYpn, which is a deviation of the component center with respect to the rotation center of the spindle 44n. When rotating 44 by α degrees, the position of the moving base 24 is corrected by the amount of displacement dXαn, dYαn and mounted. The displacement amounts dXαn and dYαn are the same as in the case of FIG.
dXαn = ΔXpn−Lcos (θ + α), dYαn = ΔYpn−Lsin (θ + α),
L ² = ΔXpn ² + ΔYpn ² , tanθ = ΔYpn / ΔXpn
It can be calculated from

DESCRIPTION OF SYMBOLS 11 ... Base, 12 ... Board | substrate conveyance apparatus, 13 ... Component supply apparatus, 15 ... Component recognition camera, 17 ... Upper end member (support member), 23 ... Control apparatus, 24 ... Moving stand, 25 ... Camera for board recognition, 26, 40 ... Component transfer device, 29 ... Suction nozzle (component sampling unit), G, Ga ... Reference gauge, Gm ... Reference mark, Gm1, Gm2 ... First and second reference marks, S ... Substrate, P ... Components , O1 ... substrate camera optical axis, O2 ... component camera optical axis, O3 ... suction nozzle center line, 42 ... rotary head, 44 ... spindle.

Claims (1)

Supplied by a substrate transport device provided on the base for carrying in / out and positioning and holding the substrate, a movable base supported so as to be movable in two directions of the X direction and the Y direction with respect to the base, and a component supply device A component having a rotary head that is capable of indexing and rotating a plurality of spindles that are collected and mounted on the substrate that is positioned and supported on the substrate transfer device, and is detachably attached to the movable table In an electronic component mounting apparatus including a transfer device, a substrate recognition camera fixed to the moving table, and a component recognition camera fixed to the base, the reference mark provided on the base is the component mark. An image is picked up by a recognition camera, the moving base is stopped at a predetermined position with respect to the coordinate origin, and a reference mark having the same or constant positional relationship as the reference mark is taken by the substrate recognition camera. And the positional relationship between the optical axis of the substrate recognition camera and the reference mark measured based on the position of the image of the reference mark imaged by the substrate recognition camera, and the image captured by the component recognition camera. Calibrating the coordinate position with respect to the coordinate origin of the optical axis of the component recognition camera based on the positional relationship between the optical axis of the component recognition camera and the reference mark measured based on the position of the image of the reference mark, The moving base is moved to a predetermined position so that the center of rotation of the spindle indexed at the mounting point is positioned at the coordinate position of the optical axis of the calibrated component recognition camera, and each spindle is moved to the first rotation angle. The tip of the suction nozzle mounted on each spindle is imaged with the component recognition camera while being positioned at a second rotation angle position different from the position and the first rotation angle position. The center position of the tip image of each suction nozzle at the first and second rotation angle positions indexed to the mounting point is arithmetically averaged to obtain the rotation center position of each spindle, and obtained from the image of each spindle. A correction method for an electronic component mounting apparatus, wherein each correction value of the rotation center position of each spindle is obtained from the rotation center position at the mounting point and the designed rotation center position.

Images