JP3209623B2 - Mounter - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mounter for accurately positioning and mounting a large number of chips on a substrate suitable for a contact type image sensor or the like.

[0002]

2. Description of the Related Art A contact type image sensor used for an image input device of an OA device such as a facsimile or a copying machine is shown in FIG.
As shown in FIG. 1, a large number of image sensor elements 2 such as CCDs and bipolar ICs are linearly arrayed and mounted on a long circuit board 1, and wire-bonded to a conductor layer 3 of the circuit board 1. It is hermetically sealed with a cap 4.

In such a contact type image sensor,
It is necessary to arrange the respective elements 2 on the circuit board 1 at an accurate interval, and various devices have been conventionally devised.

[0004] For example, FIG.
1 shows a method of assembling a contact type image sensor disclosed in Japanese Patent Application Laid-Open Publication No. H11-209,878. This assembling method uses an array jig 6 in which grooves 5 for accommodating a large number of pieces are arranged in a line, and the elements 2 are arranged in the grooves 5 of the array jig 6 in a face-down state, and are long. A method in which an adhesive 7 is applied to a predetermined position of a circuit board 1 having a shape like a face down and pressed against the back surface of the element 2 housed in series in the groove 5 to cure the adhesive 6 and assemble the same It is.

[0005] In this assembling method, the elements 2 can be arranged and arranged in series only by being housed face down in the grooves 5 of the array jig 6.

[0006]

However, in such an assembling method of aligning the elements 2 using the arrangement jig 6, the face surface is rubbed with the bottom surface of the groove 5 of the arrangement jig 6 at the time of alignment.
There was a risk of damaging the electrode surface and the light receiving film. In addition, when a semiconductor wafer is cut by a high-precision dicing technique to finish each piece, it is difficult to form the external dimensions completely uniformly, and the positional accuracy between the elements 2 arranged on the circuit board 1 is difficult. Was at most about ± 0.1 mm. For this reason, it has not been possible to obtain a positional accuracy of about ± several μm at which high definition and high resolution can be obtained. In addition, when arranging a large number of elements 2, variations in the individual external dimensions are accumulated, and it is difficult to automatically align the elements 2 on the array jig 6, and it is difficult to perform automatic mounting.

Here, a device in which a single element is positioned on a substrate and automatically mounted is disclosed in, for example, (A) Japanese Patent Laid-Open No. 2-9 / 1990.
4530, (b) Japanese Patent Application Laid-Open No. 1-250004, and (c) Japanese Patent Application Laid-Open No. 2-280304.

That is, the automatic assembling technology of semiconductor components uses a suction head that moves an element from a supply position to a mounting position, and a television camera that captures an image of a mounting target and detects the arrangement position and the substrate position. The movement amount or the mount position of the suction head is corrected based on the position information.

The above publication (A) discloses a method in which an image recognizing means is provided between a supply position and a mount position of an element, and an image of the back surface of the adsorbed element is taken to detect a shift amount of the element being conveyed. Further, Japanese Patent Application Laid-Open (KOKAI) No. H11-27921 discloses that an image recognizing means for picking up an image of both is provided between an element adsorbed on a mounting position and a substrate to be mounted, and a relative displacement amount between the two is detected. . Japanese Patent Application Laid-Open Publication No. H11-26086 discloses that an intermediate stage is disposed between a supply position of an element and a mounting position, and the amount of displacement is detected by image recognition means disposed on the intermediate stage. Address the misalignment.

However, the above-mentioned automatic assembling technique corrects the displacement of the mounting position by using an image recognition means including a suction head and a television camera, and can automatically mount a single element at high speed and accurately. However, a large number of elements are not aligned on a long circuit board at a predetermined interval, and individual external dimensions cannot be corrected.

Accordingly, the present invention has been made in view of the above-mentioned problems, and uses a suction head for transporting chips and an image recognizing means including a television camera to automatically mount chips at a high speed and have different external dimensions. It is an object of the present invention to provide a mounter in which even if a large number of chips or chips having slightly different external shapes are arranged, variations in the external dimensions of the individual chips are not integrated, and the chips can be mounted with an accurate spacing.

[0012]

In order to achieve the above object, a mounter of the present invention uses a chip provided with a recognition mark on its surface. Then, when aligning and mounting the chip on the circuit board, the chip adsorbed by the suction head on the mounting position and the mounted chip mounted on the substrate are imaged by image recognition means including a television camera, The positions of the recognition marks on both chips are detected,
Based on the positions of these recognition marks, the mounting positions of the sucked chips on the substrate are set. That is, the mounter of the present invention suctions a chip provided with a recognition mark on the surface, moves horizontally and vertically in conjunction with the suction head, and moves the chip from the first position to the second position. A first image recognition means including a first TV camera for imaging a chip on a table and detecting an edge position of the chip and a position of a recognition mark; a substrate on which the chip is mounted at a second position;
XY table that moves so that the position can be corrected in the direction, the position of the edge from the back surface of the chip sucked by the suction head at the second position at the upper position, and the XY table at the lower position.
A second image recognizing means including a second TV camera for simultaneously detecting an edge position of a chip previously mounted on a substrate on a table and a position of a recognition mark thereof, wherein the first and second images are provided. Based on the position information obtained by the recognizing means, the position of the recognition mark of the chip that has been sucked is aligned with the position of the recognition mark of the chip that has been mounted earlier, and XY
It is characterized in that the table is controlled and the mounting position of the sucked chip on the substrate is set. Further, in the mounter of the present invention, the chip is provided with a pair of recognition marks spaced apart on both sides of the chip surface in the arrangement direction, and the first image recognition means is provided with one edge position on one side of the chip. And the second image recognition means recognizes the edge position on one side of the sucked chip, the edge position on the other side of the chip mounted on the substrate, and the position on the other side. At the same time, and based on the position information obtained by the first and second image recognition means, the recognition mark on one side of the sucked chip and the other of the chip mounted on the substrate The position of the recognition mark on the side is aligned to control the XY table, and the mounting position of the sucked chip on the substrate is set.
Further, the mounter of the present invention is characterized in that the table at the first position is an intermediate stage in which individual chips are positioned and supplied from another chip supply position.

[0013]

In the mounted chip and the chip to be mounted, the position of the recognition mark provided on the chip surface is detected by the image recognition means, and the detected recognition mark positions of both chips are aligned on the mounting position. The planned mounting position is set based on the aligned position. Therefore, the chips that are aligned and mounted on the substrate are mounted with the intervals between the recognition marks of the respective chips set at the intervals set by the above set values, and set at predetermined intervals regardless of the external dimensions and the differences in the external shapes of the chips. Mounted.

In addition, when mounting a chip in which a pair of recognition marks are formed on both sides along the arrangement direction on the chip surface, the position of the recognition mark on one side of the mounted chip is detected, and the chip mounted on the substrate side is the other side. Are detected and the two are aligned. Therefore, even in the case of a chip having a pattern in which the outer dimensions of the mounted chip are different from the outer dimensions of the chip mounted on the substrate side, the chips are mounted at predetermined intervals between the recognition marks of each chip.

Further, when the table at the first position is set to an intermediate stage in which individual chips are positioned and supplied from another chip supply position, a shift at the time of suction at the chip supply position can be corrected, and the position by the television camera can be corrected. Recognition is assured.

[0016]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a schematic view showing one example of a mounter according to the present invention. A chip 9 such as a CCD or a bipolar IC is mounted on a substrate 8 composed of a long ceramic circuit substrate or the like.
1 shows a configuration of a mounter 10 that mounts a plurality of the motors at predetermined intervals.

FIG. 2 is a front view showing the outer shape of the chip 9 mounted on the mounter 10. Although not shown, each of the chips 9 is formed along a line from a wafer having a large number of chips 9 formed in a vertical and horizontal pattern. It is obtained by dicing in a piece shape, and a recognition mark 12 for setting a chip interval is provided on a pattern 11 on the surface of the chip 9.

The mounter 10 has a chip supply position 13 for supplying the chip 9, a first position 14 for imaging the surface of the supplied chip 9 and detecting the position of the recognition mark 12 on the chip surface, and a predetermined position on the substrate 8. Tip 9 in position
And a second position 15 for mounting the same.

Reference numeral 16 in the chip supply position 13 denotes a chip supply mechanism, in which a plurality of vertically and horizontally diced chips 9 are adhered and supported on a wafer sheet 17. Reference numeral 18 denotes a chip positioning mechanism for positioning the chips 9 attached to the wafer sheet 17, and a TV camera 19 disposed above the wafer sheet 17,
A recognizing unit 20 for recognizing the positional information of the chip 9 taken from the TV camera 19, and a motor control unit 21 for driving and controlling the XY-θ driving unit 16a of the chip supply mechanism 16 based on the control signal determined by the recognizing unit 20. It is composed of

Reference numeral 22 denotes a suction head which is disposed between the chip supply position 13 and the first position 14 and is attached to the tip of the pickup arm 23. The chip 9 positioned at the position is sucked and moved onto the intermediate stage 24 at the first position 14.

The moved chip 9 is positioned at a predetermined position by claws 25 arranged on all sides of the intermediate stage 24. This corrects the displacement of the chips 9 caused when the suction head 22 sucks the chips 9 attached to the wafer sheet 17 at the chip supply position 13.

Reference numeral 26 denotes an image of the surface of the chip 9 positioned on the intermediate stage 24, and a pattern 1 of the chip 9 is taken.
This is a first image recognition means for detecting the position of the recognition mark 12 provided on one surface, and is obtained from the first TV camera 27 and the first TV camera 27 disposed above the intermediate stage 24. And a recognition unit 28 for recognizing the position information of the inserted chip 9. That is, the recognition unit 28
As shown in FIG. 2, the distance m between the edge E of the chip 9 and the recognition mark 12 is calculated from the surface image of the chip 9 captured by the TV camera 27, and the position of the recognition mark 12 from the edge E is detected. Is done.

Next, the chip 9 at which the position of the recognition mark 12 is detected is sucked by the suction head 29 which moves horizontally and vertically in conjunction with it, and is supported on the XY table 30 at the second position 15 from the intermediate stage 24. It is moved onto the substrate 8.

Numeral 31 indicates that the suction head 29 is in the second position 1
5, the images are taken from the back side of the chip 9 sucked by the suction head 29 and the front side of the mounted chip 9 a mounted on the substrate 8, and the images of the chip 9 and the chip 9 a are taken. Recognition mark 12 and recognition mark 12a
And a pair of upper and lower reflectors 32 (32a, 32b) movably disposed between the suction head 29 and the XY table 30, and these reflectors 32 (32a). , 32b)
And a pair of upper and lower second TV cameras 33 (33a, 33b) for capturing an image of the chip 9a, and a recognition unit 28 for recognizing the chip 9 and the positional information of the chip 9a.

The recognizing unit 28 is provided with position information of the front surface image of the chip 9 picked up by the first TV camera 27 and the back surface of the chip 9 picked up by the pair of upper and lower second TV cameras 33 (33a, 33b). As will be described later, the mounting position of the chip 9 on the substrate 8 is calculated from the position information of the image and the surface image of the chip 9a, and the calculated data is given to the motor control unit 34 to provide an XY-θ driving unit. The XY table 30 is moved by a predetermined amount, and the chip 9 sucked by the suction head 29 is mounted on the substrate 8.

FIG. 3 shows the first TV in the recognition unit 28.
The camera 27 and the second TV camera 33 (33a, 33
The process of calculating the planned mounting position of the chip 9 from the position information according to b) is shown.

First, the surface of the chip 9 to be mounted is imaged by the first TV camera 27 at the first position 14, and as shown in FIG. The distance m between them is calculated, and the position of the recognition mark 12 from the edge E is detected. When the chip 9 is sucked by the suction head 29 and is located above the second position 15, the back surface of the chip is imaged by the upper second TV camera 33a, and as shown in FIG. Of the edge E is detected. The position detection data m from the edge E of the recognition mark 12 detected by the first TV camera 27 is taken into this, and the position of the recognition mark 12 of the chip 9 to be mounted at the second position 15 is shown by a dotted line. Is detected as follows.

At the same time, the mounted chip 9 mounted on the substrate 8 by the lower second TV camera 33b
(a), the distance n between the edge Ea of the chip 9a and the recognition mark 12a is calculated, and the edge of the recognition mark 12a of the mounted chip 9a at the second position 15 as shown in FIG. The position from Ea is detected.

The recognizing section 28 controls the two recognizing marks 12, 12.
Based on the position detection data m, n of a, both chips 9, 9
The positions of the recognition marks 12 and 12a of a are aligned,
Further, by shifting the position by a predetermined chip interval W and moving the XY table 30, the mounting position of the chip 9 is set.

Then, as shown in FIG. 3D, the substrate 8 is moved so that the position of the recognition mark 12 of the chip 9 is aligned with the position of the recognition mark 12a of the mounted chip 9a.

The chips 9 set in this way are mounted at the set chip interval W between the mounted chip 9a and the recognition marks 12 and 12a, so that an accurate distance between the chips can be obtained irrespective of the external dimensions generated by dicing. Mounted while keeping.

FIG. 4 shows another example of a chip mounted by the mounter 10.

That is, in the chip 36, a pair of recognition marks 38 and 39 are arranged on both sides along the direction in which the pattern 37 on the chip surface is arranged on the substrate 8 with a space therebetween.

At the time of mounting the chip 36, the chip 36 to be mounted has the recognition mark 38 on one side at the position p, and the mounted chip 36a mounted on the substrate 8
The position q of the recognition mark 39 on the other side is detected in the same manner as the chip 9 of the above embodiment, and the chip 36 to be mounted is aligned with the mounted chip 36a based on the detected position information. Chips 36, 36a
XY so that the recognition mark 38, 39 between
The table 30 is controlled to set the mounting position.

FIG. 5 shows the chip 36 in the recognition unit 28.
2 shows a process of calculating the mount position.

That is, the chip 36 to be mounted is imaged at the first position 14 by the first TV camera 27, as shown in FIG.
One side edge E1 of the chip 36 and one side recognition mark 38
The distance p between them is calculated, and the recognition mark 3 from the edge E1 is obtained.
8 are detected. Next, the upper surface of the chip 36 is imaged by the upper second TV camera 33a at a position above the second position 15, and the position of the edge E1 on one side of the chip 36 is recognized as shown in FIG. And
The position detection data p from the edge E1 of the recognition mark 38 detected by the first TV camera 27 is taken into this, and the position of the recognition mark 38 on one side is detected as shown by a dotted line.

On the other hand, the surface of the mounted chip 36a mounted on the substrate 8 is imaged by the lower second TV camera 33b, and as shown in FIG.
Edge E2 on the other side of chip 36a and recognition mark 3 on the other side
The distance q between the reference marks 9a is calculated, and the position of the recognition mark 39a on the other side from the edge E2 is detected.

The recognizing section 28 is provided with the recognizing marks 38 and 39a.
Based on the position detection data p, q of the two chips 36, 3
The positions of the recognition marks 38 and 39a of 6a are aligned, and the positions are shifted by a predetermined chip interval W to obtain XY.
By moving the table 30, the mounting position of the chip 36 is calculated.

Then, as shown in FIG. 4D, the substrate 8 is positioned such that the position of the recognition mark 38 on one side of the chip 36 is aligned with the position of the recognition mark 39a on the other side of the mounted chip 36a. Is moved, and the recognition mark 38 on one side of the chip 36 to be mounted is separated from the recognition mark 39a on the other side of the mounted chip 36a by a set chip interval W.

In this case, the chip 36 to be mounted is configured such that the chip interval is set by aligning the position of the recognition mark 38 on one side with the position of the recognition mark 39a on the other side of the mounted chip 36a. Even if a chip 36 having an outer dimension M different from the outer dimension L of the mounted chip 36a is used, the chip can always be mounted with the chip interval set constant.

[0042]

As described above, according to the present invention, a recognition mark is provided on a chip, a chip interval is set based on the recognition mark, and a number of chips are aligned and mounted on a substrate. With this configuration, the chips can be mounted at predetermined intervals regardless of the difference in the outer dimensions and the outer shape of the chips. Therefore, the chip can be assembled with a positional accuracy of about ± several μm, and a high-definition and high-resolution contact type image sensor can be automatically assembled at high speed.

In the present invention, in the above embodiment, the position 1 for detecting the recognition mark position on the chip surface is set on the intermediate stage where the individual chips are positioned and supplied, but is directly detected on the wafer sheet at the chip supply position. You can also. In this case, the intermediate stage can be omitted. In addition, although the TV camera of the second image recognition means uses a vertically separate TV camera, the reflection means of the optical path may be of an image switching type to be integrated. Further, the chips are arranged in a line on a long substrate, but they may be arranged in a plane in two directions of XY.

[Brief description of the drawings]

FIG. 1 is a schematic view showing an embodiment of a mounter according to the present invention.

FIG. 2 is a front view of a chip applied to the mounter of FIG. 1;

FIG. 3 is an operation explanatory diagram of FIG. 1 to which the chip of FIG. 2 is applied;

FIG. 4 is a front view of another chip applied to the mounter of FIG. 1;

FIG. 5 is an operation explanatory diagram of FIG. 1 to which the chip of FIG. 4 is applied;

6 is an exploded perspective view of a contact type image sensor to which the mounter of FIG. 1 is applied.

FIG. 7 is an exploded perspective view showing an example of a conventional mounter.

[Explanation of symbols]

 8 board 9,36 chip 10 mounter 11 pattern 12 recognition mark 13 chip supply position 14 first position 15 second position 16 chip supply mechanism 24 intermediate stage 26 first image recognition means 27 first TV camera 28 recognition unit 29 suction Head 30 XY table 31 Second image recognition means 33 (33a, 33b) Second TV camera 37 Pattern 38,39 Recognition mark E Edge

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H01L 21/52 H01L 21/60 H01L 21/68 H05K 13/00

Claims (3)

(57) [Claims] 1. A suction head for moving a chip having a recognition mark on its surface from a first position to a second position by moving horizontally and vertically, and detecting an edge of the chip and a position of the recognition mark at a first position. X is a substrate on which the chip is mounted at the second position.
An XY table for correcting the position in the Y direction, a position of an edge from the back surface of the chip sucked by the suction head at an upper position at the second position, and a chip mounted first on a substrate on the XY table at a lower position. A second image recognizing means for detecting an edge of another chip and a position of the recognition mark thereof, wherein the suctioned based on the position information obtained by the first and second image recognizing means. A mounter characterized in that a position of a recognition mark of a chip and a position of a recognition mark of another chip mounted earlier are aligned to control an XY table, and a mounting position of a sucked chip on a substrate is set. 2. A suction head for moving a chip having a recognition mark on a surface from a first position to a second position by moving horizontally and vertically, and detecting a position of an edge of the chip and a position of the recognition mark at a first position. X is a substrate on which the chip is mounted at the second position.
An XY table for correcting the position in the Y direction, a position of an edge from the back surface of the chip sucked by the suction head at an upper position at the second position, and a chip mounted first on a substrate on the XY table at a lower position. A second image recognizing means for detecting an edge of another chip and a position of the recognition mark thereof, wherein the chip is provided with a pair of recognition marks spaced on both sides of the chip surface along the arrangement direction. One edge of the chip and the position of the recognition mark on one side are detected by the first image recognizing means, and the edge of the one side of the chip adsorbed on the substrate is detected by the second image recognizing means. The other side edge of the mounted chip and the position of the other side recognition mark are detected, and one side of the sucked chip is detected based on the position information obtained by the first and second image recognition means. Recognition Mounter, characterized in that aligning the position of the other side of the recognition marks of the mounted chip to click the substrate by controlling the XY table, set the mount position on the substrate of the chip that are adsorbed. 3. The mounter according to claim 1, wherein the first position is an intermediate stage to which individual chips are positioned and supplied from a chip supply position.