JPS6025900B2 - Automatic positioning method for semiconductor pellets - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for automatically positioning a pellet when setting up a microelectronic device such as a semiconductor integrated circuit.

The automatic positioning method of the subordinate bullet will be explained using FIG. 1 and FIG.
The light passes through a transparent table 3 (hereinafter referred to as the table) which can be moved in the Y and 0 directions, but since the beret 2 is opaque, only the parts other than the beret are transmitted.

Therefore, the bullet is magnified with a magnifying lens 4, and the projection plane 5 is
Even if you make a silk image, the image of Beret 2 looks dark. In other words, the image of Beret 2 is dark and the surrounding area is bright. On the projection surface 5, as shown in FIG. It is converted into an electrical signal according to the difference in light amount caused by the difference between the blocked part and the unblocked part.

These electrical signals are electrically compared and determined based on their magnitude and the positional relationship of the light receiving section, and the table 3 is moved in the x, y, and 8 directions. By stopping 3, automatic positioning of the beret was performed. However, the size of the beret varies and is not constant, and the conventional positioning method causes problems as shown in FIGS. 3a and 3b.

In other words, Fig. 3a shows a case where the size of the pellet varies in a large direction, and even though the positions of the pellet image 6 and the light-receiving section 7 are not actually aligned, the positions are considered to be aligned electrically. Because of the judgment, the position was tilted and aligned. Third
Figure b shows a case where the bullet is scattered in a small direction. In this case, the bullet image 6 and the light receiving part 7 cannot all overlap, and the bullet image vibrates, resulting in a situation where alignment is impossible. It had become. The present invention has been devised to eliminate these drawbacks, and provides an automatic pellet positioning method for the purpose of improving the pellet positioning accuracy.

An embodiment of the automatic beret positioning method according to the present invention will be described below with reference to the drawings.

That is, as shown in FIG.
is arranged so as to fit within the beret image 6, and the two adjacent sides 19 and 20 of the beret image 6 are used as positioning sides, and these two sides 19 and 20 are provided as a pair. 1
1, 13, 14 and 16, 17 by moving the table 3 (FIG. 1) in the X, Y, and 8 directions. Next, FIG. 5 shows the positioning state when the sizes of the pellets vary, which has been a problem in the past. FIG. 5a shows the case where the size of the beret varies in the large direction, and FIG. 5b shows the case where the size of the beret varies in the small direction. As can be seen from the figure, if all the light-receiving parts are arranged so that they fit within the smallest bullet image, positioning accuracy will not be affected even if the bullet size varies.

Now, as shown in FIG.
~18 are arranged so that 10, 11, 12 are on a straight line in the On a straight line, 10, 11, 13, 14, and 16, 17 are in contact and form a pair, and 12, 15, 18 are provided apart from each other. When the beret image 6 is formed on this projection plane, positioning is performed in the following order. First, X
, the positioning in the Y direction is performed as follows. Now, the image 6 of the bellet overlaps only 12 of the light receiving sections 10, 11, and 12 in the x direction, and the images do not overlap on 10 and 11, and the light receiving section converts these states into electrical signals. and convey it as information.

In other words, when light hits the two on the right among the three light receiving sections in the It works to move it in the direction of the middle arrow A). In addition, the light receiving units 10, 11, and 12 constantly monitor the image 6 of the moving beret, and when the image of the beret overlaps 11 and 12 but does not overlap 10, it is determined that the position has been achieved, and the table 3 stops.

Positioning in the Y direction is also performed at the same time, and the determination and operation in the positioning direction are the same as in the x direction. In state a, Y
Beret's image 6 is superimposed on only 15 of the direction light receiving sections 13, 14, and 15, and no images are superimposed on 13 and 14. Therefore, it works to move the beret image 6 upward (in the direction of arrow B in the figure). The method for determining whether to stop is the same as in the x direction, and the beret image 6 overlaps the light receiving sections 14 and 15, but it stops when it does not overlap the image 13. This completes the positioning in the x and y directions as shown in b. Next, positioning in direction a is performed.

In the state b in the figure, the beret image 6 overlaps only 18 of the light receiving parts 16, 17, and 18 in the 0 direction, and 16, 1
It does not overlap with 7. The light receiving section converts these states into electrical signals and transmits them as information. The converted information is electrically compared and determined, and then the beret image 6 is caused to rotate to the right (rotation in the direction of arrow C in the figure). In addition, the light receiving units 16, 17, and 18 constantly monitor the image 6 of the pellet during operation, and when the image 6 of the pellet overlaps with the light receiving units 17, 18 but does not overlap with the light receiving unit 16, it is determined that the positioning has been completed. It is determined that the rotation of the table 3 is stopped. When the 0 direction moves, the positions in the x and y directions naturally shift, so
, and the Y direction as well, positioning is performed each time so that the position does not shift.

This completes the positioning in the x, Y, and 0 directions as shown in c. As described above, the present invention contributes to the automation of assembly of semiconductor integrated circuit devices because even if the sizes of the pellets vary, accurate positioning can be performed based on two adjacent sides.

[Brief explanation of the drawing]

Fig. 1 is a schematic diagram of the subordinate positioning device, Fig. 2 and Fig. 3 a, b are diagrams each explaining the method of positioning the submissive, Fig. 4, Fig. 5 a, b, and Fig. 6 a. , b, and c are diagrams each explaining the positioning method of the present invention. 1... Floodlight, 2... Berets, 3.
...Tefle, 4...Magnifying lens, 5.
...Projection plane, 6...Berrett image, 7...
... Light receiving section, 10, 11, 12.... ...X-direction positioning light receiving section, 13, 14, 15. ．．・．． ．． - Y-direction positioning light receiving section, 16, 17, 18... 8 directions positioning gate light receiving section, 19, 20... positioning side. Eagle - Figure 2 Figure 3 Ticket 4 Figure Nurturing Figure 6

Claims (1)

[Claims] 1 In an automatic semiconductor pellet positioning method, the amount of light generated by a semiconductor pellet image is converted into an amount of electricity, and the semiconductor pellet is moved depending on the amount of electricity to position the semiconductor pellet. Position detecting sections each having at least three light receiving sections are provided in the X direction, the Y direction, and a direction orthogonal to at least one of these, and two of the three light receiving sections of each position detecting section are mutually arranged. By connecting them to form a pair and placing the other one away from the pair, the semiconductor pellet can detect one of the two light-receiving parts of the pair using at least three position detection parts. 1. A method for automatically positioning semiconductor pellets, characterized by positioning the semiconductor pellets so that the two pellets are on the outside and the remaining two are on the inside.