JPH07332955A - Inspection of mounted printed board - Google Patents

Abstract

PURPOSE:To inspect the defective soldering by forming the intense and weak pattern of the luminance of a surface based on the amount of the received reflected light with regard to the inspection method, which scans upper surface of a mounted printed board with fine beam light, receives the reflected light along the irradiated optical axis from the irradiated position with a photoelectric transducer, and obtains the shape of the surface. CONSTITUTION:The fine beam light is generated from a light source 1. The light is made to scan approximately vertically on a mounted printed board 6 through a polygon mirror 3 and a light projecting ftheta lens 5. The reflected light in the direction along the optical axis of the irradiation with the fine beam light from the irradiated position is detected with a photodiode 10. The output of the photodiode 10 is binary-coded, and the two-dimensional pattern of the strong intensity and weakness of the luminance is formed, and the shape of the surface is obtained. Thus, the soldering is inspected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for inspecting a mounted printed circuit board which inspects a solder shape by irradiating a mounted printed circuit board with a fine beam of light and receiving the reflected light. is there.

[0002]

2. Description of the Related Art In recent years, inspection of a mounted printed circuit board for component misalignment, defective parts, defective solder, etc. is performed by irradiating a narrowed beam light on the mounted printed circuit board and detecting the reflected light. Some things are done by contact. Conventionally, there is a method using the principle of triangulation as such an inspection method, in which a minute beam light generated from a light source is deflected by a polygon mirror to scan the substrate, and the minute beam light from the irradiation position on the substrate is scanned. There is a method in which reflected light is received by a photoelectric conversion element, and the height of the irradiation position of the minute beam light according to the light receiving position is obtained and inspected.

[0003]

However, in the above-mentioned conventional inspection method, there is glossiness especially on the solder side,
Moreover, since it is often inclined, the reflected light from the irradiation position of the minute beam light may not enter the photoelectric conversion element. For this reason, there is a problem that the photoelectric conversion element cannot detect it, an accurate height is not required, and a solder inspection cannot be performed.

Therefore, according to the present invention, the reflected light along the irradiation axis of the minute beam light to the substrate is received by the photoelectric conversion element, and the tilted state of the inspection surface is detected from the magnitude of the received light amount to obtain the solder shape. Is to be able to inspect.

[0005]

In order to solve the above-mentioned problems, a method of inspecting a mounted printed circuit board according to the present invention scans the mounted printed circuit board with a minute beam light, and detects the position from the irradiation position of the minute beam light. A method of receiving reflected light along an irradiation optical axis of a minute beam light with a photoelectric conversion element and obtaining a solder shape, in which a measurement region to be inspected is divided into a plurality of small regions, and the photoelectric conversion device is used. The surface shape is obtained by binarizing the brightness of each of the small areas from the light amount of the received reflected light and creating a two-dimensional pattern of the brightness intensity.

[0006]

According to the above method, the brightness is detected for each of the divided areas from the light quantity of the reflected light received by the photoelectric conversion element,
The brightness is binarized to create a two-dimensional pattern of brightness. By detecting the inclination state of the substrate surface from the two-dimensional pattern of brightness, the solder coating is inspected for goodness or badness.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows an inspection apparatus for a mounted printed circuit board to which the present invention is applied. In FIG. 1, reference numeral 1 is a light source for generating a minute beam of light with which the mounted printed circuit board 6 is irradiated. Reference numeral 2 is a collimating lens system for converging the minute light beam from the light source 1 into a parallel light beam. Reference numeral 3 is a polygon mirror for deflecting the parallel light flux and deflecting the reflected light from the mounted printed circuit board 6. A polygon motor 4 drives the polygon mirror 3 to rotate. Reference numeral 5 denotes a light projecting fθ lens that collects the parallel light beam deflected by the polygon mirror 3 and irradiates the mounted printed circuit board 6 substantially vertically. Reference numeral 6 is a mounted printed circuit board to be inspected.

Reference numeral 7 indicates a portion of the reflected light from the irradiation position of the minute beam light on the mounted printed circuit board 6, which has returned along the light projection optical axis through the light projecting fθ lens 5 and the polygon mirror 3. It is a tunnel mirror that deflects reflected light in a direction perpendicular to the printed circuit board 6. Reference numeral 8 is a lens for collecting the reflected light deflected by the tunnel mirror 7. Reference numeral 9 denotes a diaphragm that blocks reflected light from directions other than the vertical direction. Reference numeral 10 denotes a photodiode that receives the reflected light in the vertical direction and converts the received light amount into an electrical output.

Reference numeral 11 is a table for fixing the mounted printed circuit board 6. A ball screw 12 moves the table 11 in the sub-scanning direction (arrow y direction) by rotating. A ball screw motor 13 rotates the ball screw 12. Reference numeral 14 is a guide rail for guiding the table 11. The operation of the mounted printed circuit board inspection device configured as described above will be described.

The minute light beam generated from the light source 1 becomes a parallel light beam by the collimator lens system 2, passes through the holed portion of the tunnel mirror 7, is deflected by the polygon mirror 3, and is condensed by the light projecting fθ lens 5. The minute beam light is irradiated as an optical axis on the mounted printed circuit board 6 substantially vertically. At this time, the minute beam light generated from the light source 1 is
As the polygon mirror 3 rotated by the polygon motor 4 rotates, the mounted printed circuit board 6 is scanned in the main scanning direction (arrow x direction) in the figure.

Reflected light reflected from the irradiation position of the minute beam light in the direction of the optical axis of the minute beam light irradiation (substantially vertical direction) passes through a light projecting fθ lens 5, a polygon mirror 3, a tunnel mirror 7, a lens 8 and a diaphragm 9. Through the photodiode 1
Lead to zero. At this time, the reflected light in the vertical direction is the projected light f.
The reflected light is condensed by the θ lens 5 and the lens 8 and is received by the photodiode 10. Further, a stop 9 provided between the lens 8 and the photodiode 10 blocks light other than the reflected light in the optical axis direction of the minute beam light irradiation. Therefore, only the light amount of the reflected light reflected from the irradiation position of the minute beam light in the optical axis direction of the minute beam light irradiation can be accurately measured. Hereinafter, the amount of light received by the photodiode 10 (reflected light in the direction of the optical axis of the minute beam light irradiation) will be referred to as incident brightness.

Since the photodiode 10 receives light reflected in the vertical direction from the mounted printed circuit board 6, when the inclination of the solder surface is gentle or when no solder is attached,
Since the reflected light in the vertical direction increases, the output increases,
On the other hand, when the inclination of the solder surface is steep, the reflected light in the vertical direction is small and the output is small. Utilizing this feature,
As an embodiment of the present invention, a method of inspecting a solder defect by a pattern of reflected light intensity, which is applied to the mounted printed circuit board inspection apparatus, will be described below.

First, a measurement area is set in advance prior to the inspection. FIG. 2 shows an example of setting the measurement area for solder inspection. FIG. 2A shows a plan view of a solder connection portion of a mounted component to be inspected, 21 is a lead of the mounted component, 2
2 is solder, and 23 is a land on the substrate. Measurement area 2
There are two methods for setting the size in the vertical direction, which is based on the lead width as shown in FIG. 6B, and based on the land width as shown in FIG. There is a method of making the size of the land the length of the land as shown in FIGS. In this embodiment, the measurement area 24 is set as shown in FIG.

Next, the measurement area 24 set in FIG.
As shown in FIG. 3A, a plurality of vertical and horizontal divided areas 25
Split into. The size of each divided area 25 is larger than the spot irradiated by the minute beam light from the light source 1, and is set to an appropriate size that allows the photodiode 10 to sample a plurality of reflected light amount data. In the horizontal division, the width of the central divided area, which is divided into three, instead of the equal division, can be increased.

As described above, the measurement area is set to a plurality of divided areas prior to the inspection, and then the inspection is started.
The measurement area 24 is scanned with a minute light beam by the rotation of the polygon mirror 3, and a plurality of incident luminance data is sampled in each divided area 25. The characteristic value of the epi-illumination in each divided area 25 is obtained from the sampled data. As an example of how to obtain the characteristic value, there is a method in which the average of the epi-illumination in each divided area is used as the characteristic value, or the maximum value of the epi-illumination in each divided area is used as the characteristic value.

Next, this characteristic value is binarized based on a certain set value. As described above, since the epi-illumination is dark (small) on a slope and bright (large) on a plane, it is determined to be a plane if the characteristic value of each divided area is larger than this set value, and a slope if it is small. By binarizing the characteristic value of the epi-illumination in this way, for example, the result shown in FIG. 3 is obtained. Here, the shaded portion 26 indicates an inclined surface, and the blank portion 27 indicates a flat surface portion.
It is possible to recognize the solder shape by the pattern of the intensity of the divided areas.

In the present embodiment, the inspection for good or bad solder is
This is performed by detecting the solder length and the solder non-wetting shape (that is, the defective connection between the lead and the solder) from the above-described pattern of the intensity of the reflected light.

First, the detection of the solder length will be described. The solder length is measured by recognizing the connection between lead and solder, that is, the solder start end, the solder continuity, and finally the solder end by recognizing the reflected brightness pattern. To do.

The epi-illumination patterns of the three divided areas in each column in the vertical direction are (a) all slopes, as shown in FIG.
(B) two slopes, (c) two slopes with a flat central part,
(D) One slope, and (e) all are flat. A pattern in which both ends of (c) are sloped and the center is flat is called a non-wetting pattern, and the central portion is also regarded as a slope. Here, as shown by the position of the arrow A in FIG. 5, when viewed from the left end side of the measurement area 24, two divided areas are arranged in a line in the vertical direction.
If there is more than one slope, consider it as the beginning of solder. That is, it is determined that the solder is formed rightward from that position.

The range in which the start end is recognized can be set with a certain margin in order to absorb the recognition error of the lead tip position. When no margin is given to the recognition range L as shown in FIG. 6A, it is recognized whether or not the start end exists in the first column on the left end of the measurement area. On the other hand, as shown in FIG. 6B, if the recognition range L is provided with a margin and the start end A is recognized from the range of the third row from the left end of the measurement area, the slope exists in the second row. Is also recognized as the starting end. When there is no slope in the recognition range L as shown in (c), the solder length is set to 0.

When the start end of the solder is detected, as shown in FIG. 7, it is recognized whether or not the divided area in the column on the right of the divided area which is the inclined surface is the inclined surface. Be present. In the figure, it is recognized that the solder is continuous between the arrows L. As an exception, only in the central part of the non-wetting pattern in (d), if the slope on the right side is considered to be continuous.

The end of the solder is regarded as the end of the solder when the number of divided areas on the slope continuous to the adjacent row on the right is one or less. For example, the end of the solder is located at the position indicated by arrow B in FIG. It has recognized. In particular, (d) is the method of recognizing the end in the case of a non-wetting pattern as an exception. As described above, the solder length is detected by recognizing from the start end to the end.

Next, it is the recognition of non-wetting (defective) of the solder. The non-wetting shape means that the tip of the lead 21 fits in the solder as shown in the sectional views of FIGS. 9 (a)-(a). As shown in the cross section of FIG.
It means that the tip of 1 is floating. The solder non-wetting shape is
As shown in FIGS. (A)-(b) and FIGS. (B)-(b), there is a characteristic that the epi-luminance at the apex of the solder portion is high (bright). By utilizing this characteristic, the length of the apex portion (non-wetting length) is measured from the two-dimensional pattern of the intensity of the luminance shown in FIG. 3C, and if the non-wetting length is longer than the set value, it is defective. If it is short, it is judged as good. A method for obtaining the non-wetting length will be described step by step.

First, whether or not the shape is a non-wetting shape is recognized. That is, as shown in FIG. 7C, even if the brightness of the central portion of the solder is high, if the solder shape is a fan shape, it is not a non-wetting shape. Even if the non-wetting pattern shown in FIG. 4C is present, it is not recognized as a non-wetting shape if the central divided area of the column to the right of the non-wetting pattern is not a slope.

In the case of the non-wetting shape, whether it is the shape shown in FIG. 9A or the shape shown in FIG. 9B depends on the positional relationship between the measured start position of the non-wet pattern and the start end of the solder. Determine. That is, if the start position of the non-wetting pattern and the start position of the solder match, the shape of FIG.
Otherwise, the shape is determined to be (a).

After confirming the non-wetting shape, the length of the non-wetting shape is obtained. The length of the non-wetting shape is obtained from the number of consecutive non-wetting patterns. When there are a plurality of non-wetting patterns, each shape is classified and the maximum non-wetting length for each shape is adopted. For example, in the example shown in FIG. 10, the solder length L is 9 dots, and the non-wetting length is 2 dots in the shape of FIG.
(B) The maximum length of the shape can be measured as 3 dots.

The solder length and non-wetting length measured as described above are respectively compared with set values to detect solder defects. The solder length, non-wetting length, and non-wetting shape to be defective can be changed by this set value.

[0028]

As described above, according to the method for inspecting a mounted printed circuit board of the present invention, by binarizing the epi-illumination and forming a two-dimensional pattern of the intensity of the epi-illumination, there is glossiness and inclination of solder or the like. Even if the number is large, the shape can be recognized from the amount of light received by the photoelectric conversion element, and accurate solder inspection can be performed.

[Brief description of drawings]

FIG. 1 is a perspective view showing an example of an inspection apparatus to which an inspection method for a mounted printed circuit board according to the present invention is applied.

FIG. 2 is a plan view showing an example of setting a measurement area.

FIG. 3 is a plan view showing a divided area obtained by dividing the measurement area.

FIG. 4 is a classification diagram of epi-illumination patterns.

FIG. 5 is an epi-illumination pattern diagram for explaining a measurement start end of solder.

FIG. 6 is an epi-illumination pattern diagram for explaining the measurement start end of solder.

FIG. 7 is an epi-illumination pattern diagram for explaining the continuity of solder.

FIG. 8 is an epi-illumination pattern diagram for explaining a measurement end portion of solder.

9A and 9B are a cross-sectional view illustrating a non-wetting shape, a plan view of epi-illumination, and an epi-illumination pattern diagram.

FIG. 10 is an epi-illumination pattern diagram for explaining a non-wetting shape.

[Explanation of symbols]

 1 Light Source 2 Collimating Lens System 3 Polygon Mirror 5 Projection fθ Lens 6 Mounted Printed Circuit Board 7 Tunnel Mirror 8 Lens 9 Aperture 10 Photodiode 11 Table 21 Lead 22 Solder 23 Land 24 Measurement Area 25 Divided Area

Front page continued (72) Inventor Daisuke Nagai 1 8-8 Koshinmachi, Takamatsu City, Kagawa Prefecture Matsushita Judenko Kogyo Co., Ltd.

Claims (1)

[Claims] 1. A mounted printed circuit board is scanned with a minute beam light, and reflected light along an irradiation optical axis of the minute beam light is received by a photoelectric conversion element from a position where the minute beam light is irradiated, and a solder shape is formed. This is a method of acquiring, in which a measurement region to be inspected is divided into a plurality of small regions, and the brightness of each of the small regions is binarized based on the amount of reflected light received by the photoelectric conversion element. A method for inspecting a mounted printed circuit board, wherein a surface shape is acquired by creating a pattern.