JPH0399209A - Inspection apparatus for mounting board - Google Patents

Abstract

PURPOSE:To detect three-dimensional defects by guiding scattered light obtained by scanning by laser light to a pair of position detecting elements, operating luminance data and height data based on the obtained amount of light, and substituting and outputting the height data according to the luminance data. CONSTITUTION:A printed board 101 on which parts 102 are mounted is moved with a transfer means 103. The upper surface of the board 101 is scanned by laser light 106 from a laser light source 105 through a polygon mirror 107 and an ftheta lens 109. Scattered light obtained by the reflection is converged on position detecting elements 114 and 115 through reflecting mirrors 110 and 111, the lens 109 and the mirror 107. The luminance data and the height data are operated in an image operation means 118 based on the obtained amount of light. The height data are substituted according to the luminance data in an interpolating operation means 119, and the result is outputted. The data are compared with reference height data in a decision operation means 120, and the quality of the mounted state of the component is judged.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention is applicable to a mounting board inspection device for inspecting mounting defects such as misalignment of components mounted on a printed circuit board. It relied on human visual inspection.

However, as products become smaller and lighter, components on printed circuit boards are also becoming smaller and more densely packaged. Under these circumstances, it is becoming difficult for humans to maintain high inspection accuracy while mounting very fine parts for long periods of time.

Recently, there has been a strong desire to automate inspections, and an apparatus has been proposed that inspects misalignment of parts from grayscale images from a video camera.

A conventional general inspection device is shown in FIG.

In FIG. 4, 401 is a printed circuit board, 402 is a component mounted on the printed circuit board 401, 403 is a video camera, 404 Fi. An image capture circuit A/D converts the video signal from the video camera 403 and stores it in its own image memory; 405 is an edge detection circuit that detects the edge of the component from the captured image; 406 is the edge detection circuit for detecting the edge of the component from the edge information. 407 is a reference data storage memory in which the reference coordinate values of the corners of each component are stored; 4o8 is a shifter that calculates the amount of deviation of the component from the detected corner coordinates and the reference coordinates; A quantity calculation circuit 409 is an allowable deviation amount storage memory in which the allowable deviation amount of parts is stored.
Reference numeral 10 denotes a comparison/judgment circuit that compares the calculated deviation amount with the allowable deviation amount and judges whether the mounting state of the component is good or bad.

The operation will be explained below. A component 402 mounted on a printed circuit board 401 is imaged by a video camera 403, and the video signal is A/D converted by an image capture circuit 404 and captured into its own image memory. Using the image, an edge detection circuit 405 detects the edge of the component 402, and a corner detection circuit 406 detects the coordinate value of the corner of the component 402 from the edge information.

The detected coordinate values of each corner are compared with the reference coordinate values of the component stored in the reference data storage memory 407, and the deviation amount calculation circuit 408 calculates the amount of deviation between the two.

Then, the calculated amount of deviation from the standard value of the part,
A comparison judgment circuit 410 compares the tolerance value of the deviation stored in the allowable deviation amount storage memory 409 to judge whether the mounting condition is good or bad, such as deviation of parts or missing parts.

Problems to be Solved by the Invention However, in the above-mentioned inspection method, parts are imaged with a video camera and two-dimensional information is used. There is a problem in that it is impossible to detect defects such as those where the IC is mounted in a raised position or where the legs of the IC are partially raised.

In view of the above-mentioned prior art, the present invention is a mounted board inspection device that can not only inspect two-dimensional (planar) misalignment on a printed circuit board but also inspect three-dimensional defects such as floating parts easily and with high precision. It provides:

Means for Solving the Problems In order to solve the above problems, the technical solution of the present invention includes a transport means for moving a printed circuit board on which components are mounted;
a laser beam scanning means for scanning a laser beam from a laser light source onto the printed circuit board using a polygon mirror and an fθ lens; Reflecting mirrors are provided between the printed circuit board and the fθ lens at positions symmetrical to the plane, and scattered light reflecting means is configured to reflect the scattered light through the rθ lens and the polygon mirror, and the scattered light from the scattered light reflecting means is reflected by the scattered light reflecting means. A light amount detecting means that is provided with a corresponding condensing lens and a position detecting element through a pair of reflecting mirrors, and condenses the light onto the position detecting element and outputs a photocurrent signal to each of them, and a light amount detecting means that outputs the light from the light amount detecting means. Luminance data LL. of the printed circuit board and components mounted on the printed circuit board are determined by a current signal. L2
-J. - and image calculation means for calculating height data H1, H2; and brightness data L1. from the brightness data and height data from the image calculation means. L2 is Lmin ≦Ll
≦Lmax, Lmin ≦L2 ≦Lmax (
However, Lmin: any minimum brightness level, Lmax:
indicates an arbitrary maximum brightness level), the average value of height data (H1+H2)/2 is output, and the brightness data is within the range of Lmin ≦L2 ≦Lmax and Lmin
'> Ll f, or when L1>Lmax, height data H2 is output, and luminance data is Lmin ≦Ll (h
Lmax and Lmin) L2 1 tapa L2)
In the case of Lmax, output the height data Hl, and output the height data Hl.
n) Lli or L1) Lmax cutlet Lm
in>L21 or L2) Lmax, an interpolation calculation means that replaces and outputs the height data of surrounding pixels, and a combination of the height data calculated by the interpolation calculation means and predetermined reference height data. and a determination processing means for comparing the values and determining whether the mounting state of the components on the printed circuit board is good or bad.

Function The present invention scans the entire surface of a printed circuit board on which components are mounted using a laser beam, guides the scattered light obtained by reflection from the printed circuit board to a position detection means using a reflecting mirror, and scans the entire surface of a printed circuit board on which components are mounted according to the height irregularities on the printed circuit board. The condensing position of the scattered light on the changing position detection element is detected by a photocurrent signal, and the brightness data and height data of the component mounted on the printed circuit board are calculated from the photocurrent signal by the image fault calculation processing means. . Since the height data is not calculated correctly for direct light from metal parts or solder surfaces, brightness data is used to determine whether the height data is correct, and if the height data is incorrect, the height of the surrounding pixels is calculated. Replace with the data and output. By comparing the measured height data with the reference height data, you can easily and accurately inspect two-dimensional (planar) positional deviations on printed circuit boards, as well as three-dimensional defects such as floating parts. can be inspected.

EXAMPLE Hereinafter, an example of the present invention will be described with reference to FIG.

FIG. 1 is a block diagram showing an embodiment of a mounted board inspection apparatus of the present invention. In FIG. 1, 101 is a printed circuit board, 102 is a component mounted on the printed circuit board 101, 103 is a transport means for moving the printed circuit board 101, and 104 is an arrow indicating the direction of movement thereof. 105 is a laser light source, 106 is a laser beam from the laser light source 105, 107 is a polygon mirror 108, a reflecting mirror for guiding the laser beam to the polygon mirror 107, and 109 is an fθ lens. 110. 111 is a reflecting mirror 112.
113 is a condenser lens; 114 and 115 are position detection elements provided in pairs symmetrically with respect to the laser scanning direction; 116;
．． 117 is a position signal from the position detection elements 114 and 115. 118 is the position signal 116 . 117
119 is the luminance data L1, L2 and the height data H1, H2 from the image arithmetic processing means 118; An interpolation calculation means 120 calculates correct height data from H2, and a determination calculation means 120 uses the height data from the interpolation calculation means 119 to determine whether the mounting state of the component 102 on the printed circuit board 101 is good or bad.

The operation will be explained below.

Printed circuit board 101 on which components 102 are mounted
While moving the polygon mirror 1 in the moving direction 104 by the conveyance means 103, the laser beam 106 from the laser light source 105 is transferred to the rotating polygon mirror 1 using three reflecting mirrors 108.
07, polygon mirror 107 and fθ lens 109
Accordingly, the I-za light 106 is vertically irradiated onto the printed circuit board 101. Thereby, the laser beam 106 is two-dimensionally scanned over the entire surface of the printed circuit board 101.

Scattered light reflected from the printed circuit board 101 by the scanning of the laser beam 106 is reflected by a reflecting mirror 110 . 111, and are further reflected by a condenser lens 112 through an fθ lens 109 and a polygon mirror 107. Through 113, position sensing element 114
．． The light is focused on 115. Position detection element 114.11
position signal 116 from 5. 117 is output to the image calculation processing means 118. Note that in this embodiment, the position detection elements 114 . 115 Toshite PSD (Posi
tion Sensi-tive Detector
: a semiconductor position detection element) is used, and the incident position on the PSD is such that the current flowing through the electrodes at both ends of the element is inversely proportional to the distance between each electrode.

In the image calculation processing means 118, each position signal 11
6. Component 1 mounted on the printed circuit board lOl and the printed circuit board 101 from
02 height data H1. H2$- and luminance data L1,
A calculation is performed to convert it into I, 2, and the height data and luminance data are output to the interpolation calculation means 119.

The interpolation calculation means 119 determines whether the height data has been calculated correctly based on whether the level of the luminance data is within a certain range, and if it has been calculated correctly, the height data H1,
The average value of H2 is output to the determination calculation means 120. I cut it,
If the calculation is not correct, if the brightness data L1 is outside the range, the height data H2 is output, if the brightness data L2 is outside the range, the height data Hl is output, and furthermore, if the brightness data L1 and L2 are both If both are outside the range, interpolate with surrounding height data and use it as actual measured height data. Judgment means 12
Output to 0.

The determination calculation means 120 compares the measured height data calculated by the interpolation calculation means 119 with predetermined reference height data, and
This is to determine the implementation status of.

By repeating and sequentially performing the above operations, the entire surface of the printed circuit board 101 can be inspected. In addition,
This series of operations must be performed in synchronization with an appropriate signal, and in this embodiment, synchronization was achieved using a synchronization signal synchronized with the rotation of the polygon mirror 107.

Next, the image calculation processing means 118 and the interpolation calculation means 11
9 will be explained in more detail using FIGS. 2 and 3.

Image calculation processing means 118, position signal 116. 11
7 ■1 and 2 are A/D converters 201-2
04, I1 and I2 are respectively converted into digital signals and sent to a brightness calculation circuit 207. 208 to add the luminance data (
Ll) 211 and luminance data (L2) 212 are calculated.

Height calculation circuit 205. In 206, the height data (Hl) 209 and the height data (H2) are calculated using the equation (1) below based on the process 1 and the brightness data (I1+I2).
210 are calculated respectively. However, K is a coefficient for normalization.

Height data = K”I1/(I1+I2) ・・・・・・
- (1) The interpolation calculation means 119 is the image fault calculation processing means 11
Luminance data L1, L2 and height data H1,
From H2, the level of luminance data is within a certain range (LM
AX) L l) LMIN and LMAX) L
2) The brightness level determination circuit 214 determines whether the height data is calculated correctly based on whether the height data is in LMIN), and if the height data is calculated correctly, the height data H1, H2
The average value of is output to the determination calculation means 120. If the calculation is not correct, if the luminance data L1 is out of range, the height data H2 is output, if the luminance data L2 is out of the range, the height data Hl is output, and then the luminance data Ll and If both L2 are outside the range, the data is replaced with surrounding height data, which is output as actual measured height data from the height data interpolation circuit 213 to the determination calculation means 120.

In the determination calculation means 120, the actual height data H from the interpolation calculation means 119 and the reference height data from the reference height data storage memory 216 are compared in a comparison circuit 215, and the difference is outputted to the determination circuit 217 as comparison data. do. The determination circuit 217 determines the mounting state of the product based on the comparison data from the comparison circuit 215 based on the magnitude of the difference.

Next, how the interpolation calculation circuit 119 replaces the height data of peripheral pixels is shown in FIG. 3 and will be described below.

Figure 3(b). (C) shows the brightness data L1 and L2 of a component with a cross section as shown in Fig. 3 (a), but metal surfaces generally have a high brightness level, and the fillet part of the solder surface receives direct light from the PSD. The brightness level decreases because it does not return to centimeters. For this reason, height data will not be calculated correctly on metal surfaces or solder surfaces. Therefore, an arbitrary minimum brightness level L MIN and maximum brightness level L MAX are set, and when the brightness data is below the minimum brightness level or above the maximum brightness level, it is determined that the height data is incorrect, as shown by diagonal lines. . When both the luminance data L1 and L2 are incorrect, the target pixel indicated by diagonal lines in FIG. 3(d) is replaced with correct height data of surrounding pixels. Further, if either the luminance data Ll or L2 is correct, the correct height data is adopted as shown by H1 or H2 in FIG. 10(d), thereby improving inspection accuracy.

Effects of the Invention As described above, the effect of the present invention is that when measuring the height data of a component on a printed circuit board and comparing it with reference height data to determine the mounting state of the component, metal on the component can be measured. Since surfaces such as surfaces and solder surfaces reflect directly, the height data will be incorrect. In the present invention, using brightness data and height data from two directions, it is determined whether the height data is correct based on whether the level of the brightness data is within a certain range, and this incorrect data is eliminated. By replacing the height data with the correct height data of surrounding pixels, it is possible to perform highly reliable inspections without relying on human visual inspection. It is possible to inspect defects such as floating, and the automation of inspection can be promoted, which has a large effect.

[Brief explanation of drawings]

Fig. 1 is a block wiring diagram of a board inspection device according to an embodiment of the present invention, Fig. 2 is a detailed block wiring diagram of the same main part, and Fig. 3 is a conceptual diagram showing how to interpolate incorrect data in the same main part. FIG. 4 is a block diagram of a conventional board inspection apparatus. 101... Printed circuit board, 102... Parts, 103
...Conveying means, 105...Laser light source, 107...
・Polygon mirror 109...fθ lens, 110.1
11...Reflection mirror 112. 113... Condensing lens, 114. 115...Position detection element, 118
...Image calculation processing means, 119...Interpolation calculation means, 1
20... Judgment processing means.

Claims (1)

[Claims] A transportation means for moving a printed circuit board with components mounted thereon, and a polygon mirror and fθ for laser light from a laser light source.
a laser beam scanning means for scanning the printed circuit board with a lens; and scattering light obtained by being reflected from the printed circuit board by scanning the laser beam, at a position symmetrical to the scanning plane of the laser beam, and scanning the printed circuit board. Reflection mirrors are provided between the fθ lenses, and scattered light reflection means is configured to reflect the light through the fθ lens and the polygon mirror, and the scattered light from the scattered light reflection means is reflected through the pair of reflection mirrors. A light amount detection means is provided with a corresponding condensing lens and a position detection element, and focuses the light on the position detection element and outputs a photocurrent signal to each of them. Luminance data L_1 of components mounted on
An image calculation means for calculating L_2 and height data H_1, H_2, and brightness data L_1, L_2 are calculated from L_min from the brightness data and height data from the image calculation means.
If it is within the range of ≦L_1≦Lmax, Lmin≦L_2≦Lmax (where Lmin: an arbitrary minimum brightness level, Lmax: an arbitrary maximum brightness level), the average value of height data (H_1+H_2)/2 and the luminance data is Lmin≦L_2≦Lmax and Lmin>L
If _1 or L_1>Lmax, height data H_2
If the luminance data is Lmin≦L_1≦Lmax and Lmin>L_2 or L_2>Lmax, output the height data H_1, and if Lmin>L_1 or L_
1>Lmax and Lmin>L_2 or L_2>Lm
In the case of ax, an interpolation calculation means replaces and outputs the height data of peripheral pixels, and compares the height data calculated by the interpolation calculation means with predetermined reference height data, A mounting board inspection device comprising a determination processing means for determining whether the mounting state of a component is good or bad.