JPH02210875A - Method of manufacturing optical sensor array - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a method for manufacturing a photosensor array, and more specifically, it is used in an image reading section of a facsimile machine, and a plurality of photosensor chips are arranged in a straight line. The present invention relates to a method of manufacturing a photosensor array configured with.

[Prior Art] In general, when an optical sensor is used to configure a document reading section in a facsimile machine or the like, two main configurations have been known.

In one configuration, as shown in FIG. 6, a reduction optical system 2 is interposed between the original 1 and the semiconductor image sensor 3, and the reflected light containing the read data from the original 1 of the illumination for illuminating the original is transmitted to the semiconductor image sensor 3. An image is formed on the image sensor 3 and the image is read. In this case, the image sensor 3 is composed of one sensor chip consisting of a plurality of light receiving sections (hereinafter referred to as light receiving windows), and for example, reads one horizontal line of an A4 size document with a resolution of 12 pixels per 1 mm. In this case, the semiconductor image sensor must generally be configured to read 2592 pixels in consideration of the positional accuracy of the original. If the pitch of the light receiving windows corresponding to these pixels is 12 μm, the reading section (total length of the light receiving windows) of the semiconductor image sensor is approximately 31 m.
m, and the original is reduced to about 177 and read.

Although a configuration using such a reduction optical system has the advantage of not requiring a large number of sensor chips or a large sensor chip, it has the disadvantage that the device becomes larger due to the intervention of the reduction optical system. are doing.

On the other hand, as a second configuration, a sensor array is known in which a plurality of semiconductor image sensor chips each having a length of several millimeters to several 10 millimeters in the longitudinal direction are arranged to have the same size as the original to be read. In this configuration, the document and sensor array can be placed in close proximity by using an optical system that focuses the read image one-to-one, for example, a convergent light transmitter array, or they can be placed in close contact by eliminating the intervening optical system. As a result, the reading device can be made compact.

In this case, as shown in FIG. 7, semiconductor image sensor chips Sll to
5INIS21 to S2N are arranged in a staggered manner, and as shown in FIG.
``' There is a configuration in which S s are arranged in a straight line.

In the staggered configuration, sensor chips Sll to S
Since I N and S 21 to S 2 s are spaced apart by a distance lim as shown in FIG. 7, when the document is conveyed in the direction indicated by arrow 8 in the figure, as shown in FIG. , the position where each light-receiving window of the sensor chip Sll"'SIN receives the image related to the same line of document reading, and 5
The positions at which light is received by the respective light receiving windows of 21"'52N are shifted as shown by 9 and 9' in the figure. Therefore, the reading data of the sensor chips Sll to SIN that read the original first and the sensor chips that read the original later. The read data of the chips 521 to 52N are outputted at the same output timing by using a memory.

In FIG. 8, reference numeral 6 indicates an LED for illuminating the document, 7 indicates a convergent light transmitting body array, and 5 indicates the document.

Furthermore, in order to eliminate the memory in the configuration shown in FIG. 8, as shown in FIG.
""'SIN and sensor chips 521 to 52N are mounted on semiconductor image sensor mounting boards 4A and 4B, respectively.
They may be arranged separately and read at the same position using convergent light transmitter arrays 7A and 7B corresponding to each.

The above-mentioned configuration that includes a memory or a configuration that includes two convergent optical transmitter arrays to eliminate the memory requires a large number of parts and becomes complex. On the other hand, as shown in FIG.
, ~SN are arranged in a straight line, no memory for reading is required, and there is no need for a plurality of convergent optical transmitter arrays.

By the way, in this configuration, as shown in FIG. 11 showing an enlarged view of section B in FIG. It is often placed open. For this reason,
The pitch of the light receiving windows 12 in the adjacent portion of each semiconductor image sensor chip is larger than the other light receiving window pitches a, as already shown in the figure.

Therefore, although it depends on the reading resolution, to configure a normal reading device, b≦1.58, and at most b≦28.
It is necessary to set C as small as possible in die bonding of a semiconductor image sensor chip onto a substrate.

By the way, when a plurality of sensor chips are arranged in a straight line on a substrate 4 made of ceramic or glass having a desired circuit network as shown in FIG. The adhesive is transferred to the same width and length as the sensor chip using a die bonding adhesive transfer pin, or the adhesive is applied by printing. Thereafter, a plurality of semiconductor image sensor chips are sequentially die-bonded onto the substrate 4 using a die-bonding collet.

However, in this case, as mentioned above, in order to make the distance ale extremely small, approximately 10 μm to 30 μm, a gap as shown in FIG. The adhesive 13 may get into the sensor chip and swell up on the sensor chip.

In this case, if the adhesive is conductive such as Ag paste, a short circuit may occur in the circuit patterns arranged on the surfaces of the image sensor chip S and 5k-1. Even if the adhesive is not conductive, the adhesive may reach the light receiving window 12 at the end of each sensor chip and adversely affect its photoelectric conversion characteristics.

As a general method to avoid this, for example,
-231757, that is, by setting the angle θ of the mutually adjacent end faces of each sensor chip 5koSk-1 to 90" or less as shown in FIG.
A method has been proposed to prevent the adhesive 13 from rising onto the sensor chip.

[Problems to be Solved by the Invention] However, in this method, a special configuration must be used when cutting the semiconductor image sensor chip into chips from the wafer.

In other words, a dicing saw is used to cut chips from semiconductor chips in the film, and to cut at an angle, the diamond blade of the dicing saw is operated at a speed of 30,000 to 35,000 per minute. Modification of the position of the spindle for rotation or a dedicated spindle is required.

In other words, as shown in FIG. 14, it is necessary to modify the spindle 19 on which the diamond blade 20 is attached by tilting it by an angle θ2 with respect to the cutting table 18 on which the semiconductor wafer 17 is mounted and attaching it to the dicing saw body. . Further, a dedicated dicing saw device may be required.

As a result, the cost of manufacturing chips increases due to the need to modify dicing saws that require precision or use specialized dicing saws, and once a dicing saw has been modified, it is difficult to maintain its accuracy. In some cases, it could no longer be used for normal wafer cutting.

The present invention has been made in view of the above-mentioned conventional problems, and its purpose is to prevent adhesive from building up on the sensor chip without requiring modification of the dicing saw or a dedicated dicing saw. An object of the present invention is to provide a method for manufacturing a photosensor array that makes it possible to do this.

[Means for Solving the Problems] To this end, the present invention provides an optical sensor in which a plurality of optical sensor chips each having a plurality of optical sensors that perform photoelectric conversion arranged in a predetermined direction are bonded to a support. The method for manufacturing an array is characterized in that the length of the area where adhesive is applied for adhesion in the direction in which the optical sensor chips are arranged is shorter than the length in the direction in which each of the plurality of optical sensor chips is arranged. .

[Function] According to the above configuration, when bonding the sensor chip to the support, it is possible to prevent the adhesive from extruding from the gap between adjacent sensor chips and rising onto the sensor chip.

[Example] Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 1 shows one embodiment of the present invention, and is a top view of a substrate as a support on which an optical sensor array is arranged. In the figure, 21 is a ceramic substrate on which a semiconductor sensor chip is disposed. P, ~PH are ceramic substrates 2, respectively.
This is the area where an adhesive is applied when bonding the sensor chip on the sensor chip by die bonding. It is shorter than the length plus the gap pressure @c shown in FIG. 11. Also, in Figure 1, l,
is defined as fl 2- (L-Jl 1)/2, and this length jZ2 is determined depending on the length and the characteristics of the adhesive used.

In this embodiment, the die bonding adhesive is transferred onto the substrate 4 using a stamping bottle 22 shown in FIG. By designing the stamping to be approximately 0.1 to 0.2+o++ shorter than the stamping, it is possible to prevent the spread of the adhesive during transfer from exceeding the length of the application area, jZ. In addition, the width of the stamping bottle 22 is set to be about 0.2 mm shorter than the width of the sensor chip, and the thickness of the adhesive transferred by the stamping bottle 22 is set to 30 to 70 μm, with a maximum thickness of 100 μm.
Do not exceed μl.

The adhesive used in this example was 84-ILMI-51 (trade name: Nihon Aplestick Co., Ltd.), but it is limited to this adhesive, especially if the viscosity of the adhesive is about 8,000 to 12,000 cps. isn't it. Moreover, the pressure when die-bonding each image sensor chip onto the transferred adhesive is set so that the thickness of the adhesive layer after die-bonding is 10 to 20 μm.

Furthermore, the size of the semiconductor image sensor chip used in this example is 7.8 mm x 0.7++v~35.
0II1m×1.5mm.

A configuration in which stable die bonding without adhesive swelling under the above conditions was obtained by setting the aforementioned length 12 to about 1 mm.

FIG. 2 shows a cross-sectional view of an image sensor array constructed by die-bonding a plurality of semiconductor image sensor chips using the method of this embodiment. S, -SN are semiconductor image sensor chips each having a light receiving window corresponding to a pixel of a read image, and P1' -p, I are adhesives.

FIG. 3 is an enlarged view of section A in FIG. 2, and as is clear from the figure, no swell of adhesive has occurred on the sensor and sa chips.

In the above embodiment, the adhesive was applied by transfer using a stamping bottle, but there is also a method of applying the adhesive by printing.

FIG. 5 shows the printing mask 23 in this method, and in the same figure, L is the length of each semiconductor image sensor chip plus the distance between the chips, as in the above embodiment, M1 to MN are meshes for printing adhesive, respectively. In the figure, 14 is designed to have the same size as the length of the adhesive application area in the above embodiment, but j2s, which is the length of the application area located at both ends of the mask 23, is i
It is longer than 4. This is because the length ILS is the length of the sensor chip die bonding area at both ends, so there is no need to shorten the adhesive application area in the outer part.

This example does not use a stamping bin, so
It becomes possible to control the area and the like in adhesive application more accurately than in the embodiments described above. In particular, since the coating thickness of the adhesive can be accurately controlled, adhesive extrusion due to pressure during die bonding is further reduced, and as a result, there is a greater variety of adhesives to choose from than in the previous embodiments.

[Effects of the Invention] As is clear from the above description, according to the present invention, when bonding a sensor chip to a support, the adhesive leaks from the gap between adjacent sensor chips and bulges onto the sensor chip. It becomes possible to prevent

As a result, it is possible to eliminate the influence of the raised adhesive on the light receiving part of the optical sensor, and it is possible to eliminate electrical interference.

It has become possible to manufacture optically superior optical sensor arrays.

Further, by using this method, factors that increase costs such as modification of the dicing saw can be eliminated, and furthermore, this method can be easily introduced into the manufacturing process.

[Brief explanation of the drawing]

FIG. 1 is a schematic top view showing an embodiment of the present invention and shows the adhesive application area of sensor chips arranged on a substrate. FIGS. 2 and 3 show the substrate shown in FIG. FIG. 4 is a schematic top view of a stamping pin used in the embodiment shown in FIG. 1, and FIG. 5 is a schematic diagram of a mask used in another embodiment of the present invention. A top view, FIG. 6 is a schematic perspective view showing an example of the configuration of an image reading device, FIG. 7 is a top view showing an example of the sensor chip arrangement, and FIGS. 8 and 9 are the arrangement shown in FIG. 7. Fig. 10 is a top view showing an example of the sensor chip arrangement; Fig. 11 is a partially enlarged view of Fig. 1O; Fig. 12 is a schematic side view of the arrangement shown in Fig. 11. Side view for explaining the problem FIGS. 13 and 14 are side views showing a conventional solution to the problem shown in FIG. 12. 21... Mounting board, 22... Stamping pin, S, %SN... Semiconductor image sensor chip, PI~
PN...Adhesive application area, p, I~PN'...Applied adhesive, M1~
M,...print mask mesh, 23...mask. Figure 1 Figure 2 Figure 3 Figure Figure Figure Figure N-1 Figure 10 Figure 11

Claims (1)

[Scope of Claims] 1) A method for manufacturing a photosensor array in which a plurality of photosensor chips each having a plurality of photosensors that perform photoelectric conversion arranged in a predetermined direction are formed by bonding them to a support, comprising: A photosensor array characterized in that the length of a region to which an adhesive for adhesion is applied in the arrangement direction of the photosensor chips is shorter than the length of each of the plurality of photosensor chips in the arrangement direction. manufacturing method. 2) The optical sensor according to claim 1, wherein the optical sensor chip and the adhesive application area corresponding to the optical sensor chip are adhered so that their centers in the arrangement direction coincide with each other. Array manufacturing method. 3) Among the plurality of light sensor chips, the light sensor chips other than the light sensor chips located at both ends in the array and the adhesive application area corresponding to each of the light sensor chips are different from each other in the array. 2. The method of manufacturing a photosensor array according to claim 1, wherein the bonding is performed so that the centers in the directions coincide with each other.