WO2006129873A1

WO2006129873A1 - Substrate inspection device and inspection method

Info

Publication number: WO2006129873A1
Application number: PCT/JP2006/311377
Authority: WO
Inventors: Yoshio Wakita; Koichi Rikitake; Yuki Kitamura; Shinya Umetani
Original assignee: Tecnos Co., Ltd.
Priority date: 2005-06-01
Filing date: 2006-05-31
Publication date: 2006-12-07
Also published as: TW200704912A; TWI289658B; JP2007013112A

Abstract

It is possible to rapidly and accurately evaluate a pressurized connection between terminals via an anisotropic conductor. An overlap region (10) between terminals (11, 12) arranged in parallel to a substrate (1) and a TCP (2), respectively is divided into a plurality of regions A, B, C, ... along the parallel direction and region images 20a, 20b, 20c, ... of the respective regions A, B, C, ... are obtained. When acquiring the images, the light irradiation and acquisition of the image (20) is performed in the oblique direction with respect to the overlap region (10) of the terminals (11, 12) in parallel on the surface opposite to the side where the terminal (12) of the TCP (2) is arranged and the light quantity applied tot he respective regions A, B, C, ... and the distance to imaging means (16), and the photo sensitivity of the imaging means (16) are made identical in all the regions A, B, C, .... An output signal (30) in the parallel direction corresponding to the respective images (20a, 20b, 20c, ...) is obtained and by utilizing the characteristic that the output signal (30) expresses a convex/concave shape, evaluation is made depending on whether a numeric value selected from the peak position (31), the amplitude (32), the frequency (33), the half band width (34) exceeds a reference value for each of the images a, b, c,... corresponding to the terminal. With this configuration, it is possible to rapidly and accurately evaluate the pressurized connection state.

Description

Specification

Substrate inspection device and inspection method

Technical field

The present invention relates to a TCP (Tape Carrier) connected to a terminal of an electronic device board, for example, a printed circuit board (hereinafter referred to as PWB) via an anisotropic conductor. Package, tape carrier package, claims and “TCP” includes “FPC”, “TAB”, and “COF” described below), etc. The present invention relates to a substrate inspection apparatus and an inspection method for accurately and rapidly inspecting a conduction state of a connection portion with a terminal.

[0002] Here, the "FPC (Flexible Printed Circuit)" exemplified as "TCP" refers to that used for an interface with a PWB or COG substrate on which a liquid crystal driver IC is mounted. The “TAB (Tape Automated Bonding)” refers to a kind of IC chip mounting technology used for TCP. The “CF (Chip On Flexible Printed Circuit Board)” refers to a system in which an IC chip is directly connected to an electrode terminal on a flexible printed circuit board.

Background art

[0003] Liquid crystal display devices, plasma display devices, and the like are each manufactured by connecting and mounting a driving Ic chip or the like on a panel board. When connecting or mounting the IC chip or the like to the panel substrate, the opposing terminals of the substrate and the IC chip are connected in a conductive state. Anisotropic Conductive Film (hereinafter referred to as “ACF”) or an anisotropic conductor such as an anisotropic conductive adhesive is used.

ACF used as an anisotropic conductor is a sheet in which an ACF particle is dispersed and an adhesive layer is laminated on the surface of a binder layer. ACF exhibits the thermocompression bonding function by applying heat to melt the binder layer material and then curing it.

[0004] Generally, the anisotropic conductor is attached to the substrate side or the chip side in advance, and the parallel terminals provided on the panel substrate or PWB are opposed to the parallel terminals provided on the TCP. After positioning the corresponding terminals (electrodes) to face each other, add an ACF thermocompression bonding machine, etc. The pressure tool presses and heats the overlapping area between the terminal of the board and the terminal of TCP.

[0005] By this pressing and heating, the conductive particles in the anisotropic conductor are pressed between the opposing terminals, and both the opposing terminals are electrically connected via the conductive particles (FIG. 15). reference).

Therefore, in the thermocompression bonding process, it is necessary that ACF is uniformly filled between the terminals of the substrate and the TCP terminals, and that the terminals be crimped at an appropriate pressure, temperature, and time.

[0006] However, there may be a case where the adhesive cannot be uniformly filled or crimped due to excessive residual adhesive between the terminals, generation of bubbles or the like, peeling of the conductive film, and the like. Product defects in the mounting process of liquid crystal display devices and plasma display devices include this type of material failure and crimping failure between terminals due to foreign matter adhering to the substrate.

Furthermore, the connection state by ACF depends on the ACF curing characteristics that change with time, the temperature, pressure, pressurization time, pressure distribution, etc., adjusted for the limited tact time. Since the parameters are intertwined with each other, optimization of the parameters is not easy and causes product defects in the mounting process.

[0007] When TCP2 is mounted on the liquid crystal substrate 1 ', the inspection of the connection state between the two terminals is different from that of each terminal (lead) 11 of the liquid crystal substrate 1' as shown in FIG. Via conductive conductor 3

′ Is taken from the back side of the liquid crystal substrate 1 and the overlap region with the terminal 12 of the connected IC is taken. Based on the output signal 30 in the area of the electrode portion of the photographed image, the overlap region is 10) Calculate the image area of the conductive particles 4 within 10 (the image area due to the indentation 5 generated by the conductive particles deforming the electrode), and whether the calculated image area exceeds the preset reference value. In general, it is determined whether or not the number of images that have passed the determination exceeds a reference number within a preset area (see, for example, Patent Document 1).

[0008] However, when TCP is mounted on a printed circuit board! / The current situation is that the inspection of the connection state between the two electrode terminals is still focused on visual inspection. This is because the printed circuit board is not transparent like the liquid crystal substrate 1 ′, and as shown in FIG. 15, the image of the indentation 5 due to the conductive particles 4 cannot be obtained from the back side force of the liquid crystal substrate 1. It is. [0009] When the inspector visually inspects, as shown in FIGS. 16 (a) and 16 (b), the overlapping area of TCP2 and the printed circuit board 1 is irradiated with light in a direction oblique to the TCP2 side force. The state of the unevenness generated on the TCP2 film in the overlapping region 10 by the reflected light is confirmed by the eye of the inspector.

TCP2 has a terminal 12 on a film that is also made of polyimide resin and the like, and after it is thermocompression-bonded with ACF3, it is cooled, and the adhesive layer shrinks when it is cooled. (See FIG. 2, which is an illustration of an embodiment of the present invention).

The degree of unevenness reflects the pressure, temperature, and time at the time of crimping, and with good crimping, the unevenness of the part where the terminal is present and the part where it is not present appears remarkably, and the period of the unevenness is constant, and the crimping is not good. However, the gradient near the boundary between the concave and convex portions becomes gentle, and the irregularities are not remarkable, and the periodicity of the irregularities tends to be disturbed. .

[0010] Since the inspection method that the inspector visually inspects is a sensory inspection based on experience, there is a problem that only an ambiguous evaluation determination can be made. In addition, if a skilled inspector does not see it, fine defects cannot be found, the standards do not match between inspectors, labor is required, the inspection time is long, and the inspector fatigues. There is a problem that it is difficult to detect defects.

[0011] As described above, when the reliability of the inspection result is low, it is difficult to immediately feed back the inspection result to the AC F thermocompression bonding machine of the mounting apparatus that is the upstream equipment, and to immediately take measures against the process loss. ,. In addition, when the inspection work and inspection area are large, such as a panel board for a liquid crystal display device of a large TV or a panel substrate for a plasma display device, the operation of moving the object to be inspected smoothly to the inspection location There is a problem that it takes time to move the substrate, it is difficult to inspect the board, and it is difficult to prevent adhesion of dust and particles (particles).

On the other hand, in Patent Document 2, a sheet-like film (hereinafter referred to as a “polarizing film”) on which a plurality of films such as a polarizing film are bonded together, or minute irregularities on the surface of the polarizing film or foreign matter mixed inside the polarizing film A technique for detecting and inspecting foreign matter in a film having a function of detecting the like is disclosed.

[0013] This apparatus uses a photographed image detected by defocusing an image on the surface of a polarizing plate film. , Smoothing processing, second-order differentiation processing, and binary density processing are performed, and uneven defects and the like are detected based on the threshold value for extracting the bright and dark parts corresponding to the defective portion.

Patent Document 1: Japanese Patent Laid-Open No. 2004-342687

Patent Document 2: JP-A-6-235624

Disclosure of the invention

Problems to be solved by the invention

By the way, the mounting state between the printed circuit board 1 and the TCP 2 is determined in the same manner as in the Patent Document 1 or the Patent Document 2 described above, with illumination means (light source) 7 and photographing means (imaging device for acquiring images). ) A method of evaluating and inspecting with images using 6 etc. is also conceptually conceivable. For example, from the oblique direction indicated by the arrow in FIG. 16 (b), the situation of unevenness generated on the film surface of TCP2 in the overlap region 10 is acquired as an image, and based on the image for each area corresponding to each terminal, A method for inspecting the unevenness of each area, that is, the crimping state between the terminals 11 and 12, by using an output signal obtained from the image can be considered. This is because the degree of reflection of light in a specific direction differs depending on the uneven shape of each part, and this is reflected in the image luminance or the distribution of output signals obtained by the image force.

[0015] However, the image acquired in this way! The distances from the position of the photographing means 6 are different between the terminals arranged in parallel in one image and the film uneven surface on each terminal. Therefore, it is difficult to focus on all terminals and their film surface. In addition, even for a single terminal, the difference in distance occurs depending on the part, and the expected angle of the camera differs at each position, so it is difficult to focus on all parts of the terminal. is there. It is not possible to objectively evaluate the image brightness of the part that is out of focus and the image brightness of the part that is in focus using the same criteria.

Further, for example, even if the concave and convex cross-sectional shape is the same among a plurality of terminals, the image luminance depends on the difference in distance from the photographing means 6 or the difference in distance from the illumination means 7, etc. Appears in different shades. Specifically, as shown in FIG. 17, the partial force S on the right side in the figure, which is close to the photographing means 6, shows a darker color compared to the part on the left side in the figure, and the degree of shading is It can be seen that it gradually changes according to the position in the left-right direction. Therefore, the conduction state of all terminals in parallel is good. There is a problem that V cannot be objectively evaluated based on the same standard, and the level of crimping between all terminals cannot be accurately evaluated.

[0016] Further, according to the method described in Patent Document 2, it is possible to inspect the presence or absence of a defect generated in an inspection region having a flat surface such as a polarizing film. For example, as shown in FIG. When determining whether the crimping state of the terminal is good or bad based on the unevenness formation on the upper surface of CP2 (the surface opposite to the side where the terminal is provided) 13, apply that method as it is. It is not possible.

If the surface is flat like a polarizing film, the output signal also has a flat characteristic.It is easy to detect local fluctuations in the medium force, and foreign matter intervening in the inspection object, etc. The presence and position of various defects can be detected, but the quality of the crimped state between the substrate and the TCP terminal cannot be accurately determined only by this method.

Therefore, the method described in Patent Document 2 cannot accurately evaluate the quality of the crimped state of all terminals parallel to the board and TCP.

[0017] Furthermore, the board and the TCP may be warped or bent during the mounting (see FIGS. 12A and 12B, which are explanatory diagrams of the embodiment of the present invention).

If warpage or deflection occurs on the board on which TCP is mounted, the position of the output signal will deviate from the actual terminal position of the board.

[0018] For example, in area A shown in FIG. 12 (b), the board 1 on which TCP2 is mounted is slightly inclined with respect to the stage S of the board inspection apparatus, and TCP2 is implemented in area B. The mounted substrate 1 is almost parallel to the stage S. Fig. 12 (c) and Fig. 12 (d) show the cross sections in region A and region B, respectively.

Note that Fig. 12 (a) shows the board 1 on which TCP2 is mounted along the upside down direction with respect to the stage S. For ease of explanation, the above figures show the spacing between terminals. Etc. are exaggerated.

The output signal obtained from the acquired image in region A is as shown in Fig. 12 (e). The periods L and L in region A are the output signal periods L, L, and L in region B n nf lm m + l tend to be shorter than m + 2.

If TCP2 is curved in an arc as shown in Fig. 12 (a), (b), etc. For example, the states L, L,..., And L, L may all be different.

[0019] Further, the warp or deflection of the substrate 1 or TCP2 occurs, and the phase of the output signal 30 obtained from the acquired image force is shifted.

For example, warping or deflection occurs on board 1 or TCP2! /, Based on the reference waveform obtained from a good board to be inspected, the area corresponding to a pair of terminals 11 and 12 in output signal 30 If C is to be specified, the start point of region C is point E on output signal 30 and the end point is point F, as shown in Fig. 13 (a). However, point E and point F are both based on the reference waveform, and judging from the peak positions 31 and 31 of the waveform of the output signal 30, it is possible to shift the phase of the waveform.

[0020] As described above, when the phase and period of the acquired image and the output force of the acquired image are shifted, the range of the output signal corresponding to each terminal cannot be specified, and whether the crimping state between the terminals is good or bad Can't judge accurately! /.

Therefore, the present invention enables high-speed and accurate evaluation of the quality of the crimped state between the terminals of the portion where the terminal of the substrate and the terminal of the TCP are connected via the anisotropic conductor. Is an issue.

Means for solving the problem

[0022] In order to solve the above-described problem, the present invention provides a TC in an overlapping region in which a plurality of terminals are arranged in parallel.

When acquiring an image on the upper surface of P, the area is subdivided along the parallel direction of the terminals, and the output signal characteristic value and calculation are calculated based on the output signal that also obtains the image power of each divided area. The crimping state was evaluated from the value.

In this way, it is possible to objectively evaluate the output signals for the images of all the parts in the superimposed region. If the output signal can be objectively evaluated, it is possible to objectively evaluate the quality of the crimped state between the terminals, that is, the conduction state between the terminals.

[0023] Further, in order to solve the above-described problems, the present invention corrects an output signal of an acquired image based on a reference signal.

Since the range of the output signal corresponding to each terminal can be specified by the correction, it is possible to correctly determine whether the crimping state between the terminals is good or not even when the substrate or TCP is warped or bent. The invention's effect

[0024] According to the present invention, the quality of the crimped state between the terminals of the portion where the terminal of the substrate and the TCP terminal are connected via the anisotropic conductor can be evaluated at high speed and accurately.

Brief Description of Drawings

[0025] [FIG. 1] (a) shows a configuration of an inspection apparatus according to one embodiment, and (b) shows an enlarged plan view of a main part of an object to be inspected.

2] An enlarged cross-sectional view of the main part of the test object.

FIG. 3 is an explanatory view of the principle of detecting uneven portions in the substrate inspection apparatus according to the embodiment.

[FIG. 4] An explanatory diagram of inspection contents of the OK and NG parts of the substrate inspection apparatus.

FIG. 5 is a photograph showing ◯ K and NG images of the substrate inspection apparatus according to the embodiment.

FIG. 6 is a flowchart showing details of the inspection process. '

[Fig. 7] An example of conversion from an area image to an output signal is shown. (A) is a photograph showing the area image, and (b) is an output signal after conversion.

FIG. 8 is an explanatory diagram showing the configuration of the substrate inspection apparatus.

FIG. 9 is an explanatory diagram showing the relationship between the light irradiation angle, the image acquisition angle, and the unevenness of the TCP.

[FIG. 10] (a) shows a method of dividing an image area, and (b) is an explanatory diagram showing an output signal of the divided area.

FIG. 11 is an enlarged cross-sectional view of the main part showing an example of contamination.

FIG. 12 is an explanatory diagram showing the relationship between the board and TCP when warping or deflection occurs.

FIG. 13 shows signal waveforms before and after phase correction, and signal waveforms after phase correction.

[Figure 14] Output signal waveform before phase correction, reference waveform, and output signal waveform after phase correction.

FIG. 15 is an explanatory view showing a conventional inspection method.

FIG. 16A is an explanatory diagram showing a conventional inspection method, and FIG. 16B is an explanatory diagram showing a conventional inspection method.

FIG. 17 is a photograph showing an image acquired by the method shown in FIG. 16 (b).

Explanation of symbols

[0026] 1 Printed circuit board (board) 'LCD substrate (substrate)

TCP

Anisotropic Conductor (ACF) Conductive Particle

Indentation

, 16 Imaging means (imaging device), 17 Illumination means (light source) Foreign matter

0 Superimposed area

1, 11 ', 12, 12, Terminal (electrode) 3 TCP top surface

3a Convex

3b recess

4 ACF surface

5 Alignment imaging device 8 Inspection head,

9 Scan direction

0 Image (image data)

0a, 20b, 20c- ... Area image 0a ', 20b', 20c '... Area image 0, 30', 30 "Output signal 1, 31, 31" Peak position 2 Amplitude

3 cycles

4 Half width

0 Control device (controller) 1 PC (personal computer) 2 Display device (monitor) 50 Control panel

a, b, c --- image (area)

A, B, C --- region

S X— Y— Z— 0 stage

a, a ', a "Irradiation direction

β, β ', β "Image acquisition direction

θ _βΙ θ 'Irradiation angle

θ, θ 'acquisition angle

F _a Direction of illumination means

F Moving direction of the shooting means

E Start point of output signal

F End point of output signal

BEST MODE FOR CARRYING OUT THE INVENTION

[0027] As an embodiment of the above means, the following inspection apparatus can be adopted.

[0028] That is, a plurality of terminals 11 and 12; 11 'provided in parallel with the substrate 1 and TCP2, respectively.

, 12 '(a plurality of terminals 11, 12 or a plurality of terminals 11', 12 ') are overlapped with each other, and an anisotropic conductor 3 is interposed between the opposite terminals 11, 12; 11', 12 '. The opposing terminals 11, 12; 11 ′, 12 ′ are connected to each other, and the crimped state between the opposing terminals 11, 12; 11 ′, 12 ′ is determined as the terminals 12, 12 ′ of the TCP 2. In the substrate inspection apparatus that optically inspects the image data 20 of the overlapping regions 10 of the parallel terminals 11, 12; 11 ′, 12 ′ on the surface opposite to the side provided with the overlapping region 10, Divided into a plurality of areas A, B, C… along the parallel direction, and illuminating means 17 irradiates each area A, B, C ',' toward the parallel direction from the oblique direction. Then, the image data 20 of the superimposed region 10 is acquired for each of the regions A, B, C ′ · ′ by the photographing means 16, and the region images corresponding to the regions A, B, C ′,. Output signal obtained from 20a, 20b, 20c… 30 and the characteristic value calculated from the output signal 30 or the calculated value based on the calculated characteristic value by comparing with the preset reference value, the both terminals 11, 12; 11 ', A configuration for inspecting the crimp state between 12 'can be adopted. [0029] Further, the amount of light applied to each of the regions A, B, C ··· when acquiring the region images 20a, 20b, 20c ···, the regions A, B, C- And the imaging means 16, the light receiving sensitivity of the imaging means 16, the irradiation angle 0 of the light with respect to the surface direction of the substrate 1, and the image acquisition angle Θ are defined as the total areas A, B, C ′. · It is possible to adopt a configuration that is constant with '.

In this way, the quality of the crimped state between the terminals 11, 12; 11 ′, 12 ′ in the overlapping region 10 can be objectively evaluated based on the same criteria.

With this configuration, the illumination unit 17 and the imaging unit 16 are integrated, and the illumination unit 17 and the imaging unit 16 and the overlapping area 10 are relatively moved in the parallel direction. Thus, if the configuration in which the respective area images 20a, 20b, 20c ′,... Are acquired in order is used, the illumination means 17 for irradiating each area A, B, C. There is no need to provide a large number of 16s.

As a method of relatively moving the illumination unit 17 and the imaging unit 16 and the overlapping region 10 in the parallel direction, a method of moving the substrate 1 mounted with TCP2 with respect to the illumination unit 17 and the imaging unit 16, For example, a method of moving the illumination unit 17 and the imaging unit 16 fixed to each other with respect to the substrate 1 on which TCP2 is mounted, a method of moving both of them together, and the like.

Note that the illuminating means 17 and the photographing means 16 are integrated to mean that when the illuminating means 17 and the photographing means 16 are moved, the movement amounts of both are the same.

[0031] The characteristic value is a single characteristic value or a plurality of characteristic values selected from a peak position 31, an amplitude 32, a period 33, and a half-value width 34 calculated from the output signal 30, and the reference value Can be adopted for each of the areas a, b, c... Corresponding to the pair of opposing terminals 11, 12;

In this way, the level of the crimped state can be accurately evaluated for all terminals 11, 12; 11 ', 12' in parallel.

[0032] Further, in the above configuration, each of the region images 20a, 20b, 20c,... 'Is divided in a direction orthogonal to the parallel direction, and each of the divided region images 20a', 20b ', 20c' is divided. • Single or multiple characteristics selected from 'output signal 30' obtained from each and peak position 31 ', amplitude 32', period 33 ', half width 34' calculated from the output signal 30 ' It is possible to adopt a configuration in which the foreign matter 8 mixed between the opposing terminals 11, 12; 11 ′, 12 ′ is detected by a value or a calculated value based on the calculated characteristic value. If the region is divided into a plurality of regions, it becomes easier to detect local fluctuations in the output signals 30, 30 ′ due to the inclusion of the foreign object 8.

[0033] Further, if a configuration is adopted in which the output signals 30, 30 'obtained from the superposition region 10 are compared with a reference signal in the same superposition region 10 and the phase and period thereof are corrected, the correction is performed. Thus, the range of the output signal 30 corresponding to each of the terminals 11, 12; 11 ′, 12 ′ can be specified. Therefore, even when the object to be inspected is warped or bent, the terminals 11, 12; It is possible to correctly determine the quality of the crimped state between 12 '.

[0034] Further, if the photographing means 16 is an area sensor, and the acquisition width in the parallel direction in each of the region images 20a, 20b, 20c,... Can be changed according to the overlapping region 10 to be inspected. The difference in distance between each position in the single region where the image is acquired and the photographing means 16 is reduced, and further, the difference in the expected angle of the photographing means 16 with respect to each position is eliminated. For this reason, the error of the output signal 30 due to the difference in the degree of coincidence of the focus within a single region can be reduced. This acquisition width can be adjusted according to the characteristics of the object to be inspected. Also, by scanning with a narrow acquisition width, images of each of the areas A, B, C-. Therefore, the inspection speed can be greatly reduced.

In addition, the illumination unit 17 may employ a configuration including a light source having a light emission spectrum that substantially matches the light receiving sensitivity of the imaging unit 16.

In this way, the photographing means 16 can efficiently receive the light emitted from the light source to each of the areas A, B,. Therefore, the amount of light to be irradiated can be suppressed, the exposure time can be shortened, and the inspection tact can be suppressed.

[0036] Further, the irradiation angle 0 of the light by the illuminating means 17 and the acquisition angle 0 of the image by the photographing means 16 are respectively set in the TCP2 in the superposed region 10 to be inspected.

β

It is possible to adopt a configuration that can be changed according to the uneven shape of the surface on the side opposite to the side where the terminal 12 is provided.

In this way, depending on the type of the target substrate 1 or the position of the overlapping region 10, the irradiation direction of the illumination unit 17 and the image acquisition direction of the imaging unit 16 are arbitrarily set. Or, because it can be changed, it is possible to more effectively receive the reflected light from the uneven part generated on the surface opposite to the side where the terminal 12 of TCP2 is provided, and it is possible to judge whether the crimped state is good or not by the output signal 30 based on the image It becomes easy.

If the shutter speed of the photographing means 16 can be changed or adjusted, the relative conditions between the illumination means 17 and the photographing means 16 can be optimized. For this reason, it is possible to flexibly deal with various substrates 1 with different uneven pitches and height differences, and an image can be acquired effectively.

Also, if the illumination means 17 can be slid along the direction of light irradiation and the imaging means 16 can be slid along the image acquisition direction, the focus can be easily adjusted according to the unevenness of the overlapping area 10. Can be made.

[0037] It is to be noted that the board inspection apparatus configured as described above is connected to an electronic component mounting apparatus, that is, the terminals 11 and 12; 11 'and 12' of the board 1 and TCP2 via an anisotropic conductor. A configuration integrated with the electronic component mounting apparatus can also be adopted.

On the other hand, the following inspection method can be adopted as an embodiment of the means.

[0039] That is, a plurality of terminals 11, 12; 11 ', 12, which are provided in parallel with the substrate 1 and TCP2, respectively, are overlapped with each other and are different between the opposing terminals 11, 12; 11', 12 '. The opposing terminals 11, 12; 11 ′, 12 ′ are connected via the isotropic conductor 3, and the crimping state between the opposing terminals 11, 12; 11 ′, 12 ′ is Substrate optically detected by the image data 20 of the overlapping region 10 of the parallel terminals 11, 12; 11 ', 12' on the opposite surface of the TCP2 to the side where the terminals 12, 12 'are provided In the inspection apparatus, the overlapping region 10 is divided into a plurality of regions A, B, C... Along the parallel direction, and the parallel direction from the oblique direction with respect to the regions A, B, C,. The image data 20 of the superimposed region 10 is acquired for each of the regions A, B, C ′ · ′, and the regions corresponding to the regions A, B, C ′ ·· Images 20a, 20b, 20c… By comparing the output signal 30 obtained and the characteristic value calculated from the output signal 30 or the calculated value based on the calculated characteristic value with a preset reference value, the both terminals 11, 12; 11 This is a configuration of the substrate inspection method for inspecting the crimped state between 'and 12'.

[0040] Further, each of the regions A, B, C ··· when acquiring the region images 20a, 20b, 20c ··· The amount of light irradiated on each of the areas A, B, and so on. The distance between the image capturing means 16 for acquiring the image data, the light receiving sensitivity of the image capturing means 16, the irradiation angle Θ of the light with respect to the surface direction of the substrate 1, and the image acquisition angle 0 A configuration in which A, B, C... Are constant can be adopted.

[0041] The characteristic value is a single characteristic value or a plurality of characteristic values selected from a peak position 31, an amplitude 32, a period 33, and a half-value width 34 calculated from the output signal 30, and the reference value For comparison with the above, it is possible to adopt the configuration of the substrate inspection method performed for each of the areas a, b, c... Corresponding to the pair of terminals 11, 12;

[0042] Further, each of the area images 20a, 20b, 20α,... Is divided in a direction orthogonal to the parallel direction, and each of the divided area images 20a ′, 20b ′, 20c ′,. Single or multiple characteristic values selected from the output signal 30 ′ obtained and the peak position 31 ′, amplitude 32 ′, period 33 ′, half width 34 ′ calculated from the output signal 30 ′, or A configuration of a substrate inspection method for detecting the foreign matter 8 mixed between the opposing terminals 11, 12; 11 ′, 12 ′ by a calculated value based on the calculated characteristic value can be adopted.

[0043] Also, the configuration of the substrate inspection method for correcting the phase and period by comparing the output signals 30, 30 'obtained from the superposition region 10 with a reference signal in the same superposition region 10 Can be adopted.

Example

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

As shown in FIGS. 1 (a) and 1 (b), the substrate inspection apparatus of the present invention is a parallel inspection of printed circuit boards 1 of a test object in which a printed circuit board (PWB) 1 and a liquid crystal substrate 1 are connected via TCP2. And the terminals (electrodes) 12, 12, 12... Connected via the conductive particles 4 of the anisotropic conductor (ACF) 3. In the overlapping region 10, the crimping state between the corresponding terminals 11, 12; 11 ′, 12 ′ is inspected based on the image data ²⁰ .

As shown in FIG. 8, the configuration of the apparatus is as follows. The X-YZ-Θ stage S on which the substrate 1 is placed, and the substrate 1 placed on the stage S is lighted in one direction obliquely to the substrate 1. Illuminating means (light source) 17 for irradiating, an imaging device (imaging means) 16 for scanning and photographing, a control device (controller) 40, a control panel 50, and the like. In this embodiment, the imaging device 16 is an area sensor (secondary Former sensor) is adopted. It is also possible to adopt a line sensor in which only one row of photosensitive parts that receive light is arranged.

[0046] The control device 40 processes the image acquired by the imaging device 16 to produce an output signal 30, and measures various characteristic values from the output signal 30, and further calculates based on the characteristic values. Registers information on the image processing apparatus to be processed and the substrate 1 to be inspected, and on the basis of the information, the characteristic values and calculated values measured in the image processing apparatus and preset pass / fail judgment reference values And a registration / determination processing device that performs a pass / fail determination and an XYZ-Θ stage S and a control panel 50 that controls the movement of the imaging device 16 and the like. Reference numeral 41 indicates a PC (personal computer), and reference numeral 42 indicates a display device (monitor).

FIG. 1 (a) shows an inspection head 18 integrally including an imaging device 16, an illuminating means 17, and an alignment imaging device 15 for positioning, and a substrate 1 mounted on the XYZ-Θ stage S. The other components are not shown in the figure.

The alignment imaging device 15 is a camera that measures the inspection position, and the X-YZ-Θ stage S is passed through the control device 40 and the control panel 50 according to a signal from the alignment imaging device 15. Each is operated, and alignment is performed by the operation.

Using the alignment imaging device 15 and the X-Y-Z-Θ stage S, it is possible to accurately specify the measurement area of the object to be inspected.

A predetermined image can be acquired by the imaging device 16 that scans the image data 20. The condition setting of the imaging device 16 to be scanned is appropriately set by the PC 41 and the control device 40. The PC 41 can collectively manage various data transmitted from the control device 40.

In addition, as shown in l (a), the illumination means 17 can be adjusted to the optimum irradiation angle 0 by sliding in the F direction and operating in the 0 direction. Similarly, the imaging device 16 slides in the F direction.

The image acquisition angle Θ can be adjusted by operating in the Θ direction. These illumination means 17 and ø ø

The adjustment related to the imaging device 16 is appropriately set manually or by an operation key (not shown).

Further, when the image acquisition device 16 acquires an image, the control device 40 can perform shutter speed control and focus control.

[0050] Further, the image data 20 captured by the imaging device 16 to be scanned or the captured image data 20 An image obtained by processing the processed image data 20 by the image processing device and an output signal 30 obtained from the image can be displayed on, for example, a display device (monitor) 42 having a liquid crystal monitor power. Become.

In this way, after placing the test object on the -YZ- Θ stage S, all necessary areas can be inspected without touching the test object. It is designed to prevent the adhesion of particles (particles) and facilitate the inspection of large LCD panels.

[0051] The object to be inspected is a printed circuit board (PWB) 1 of a liquid crystal display device and a liquid crystal substrate 1 'connected via TCP2, and TCP2 is an electronic component for driving such as an IC chip. It is manufactured by connecting and mounting. When the TCP 2 is positioned and connected to the substrate 1 and the liquid crystal substrate 1 ′, an ACF (anisotropic conductor) 3 is employed as a connection member.

Printed circuit board (PWB) 1 terminal 11 and TCP2 terminal 12 and liquid crystal board 1 ′ terminal 11 ′ and TCP terminal 12 ′ are overlaid and connected via ACF3 conductive particles 4 respectively. ing.

[0052] FIG. 1 shows a diagram in which one printed circuit board 1 is mounted in the X direction and one printed circuit board in the Y direction. However, the number of mounted substrates 1 is not limited to the X direction and the Y direction. It is possible to support the case where one or both of these are mounted in one or both! / Ugly ladies.

[0053] For this connection, for example, after ACF3 is thermocompression bonded at 170 ° C, the material is cured by cooling, and the surface of TCP2 at the connection portion is shown in FIG. Concavities and convexities (convex portions 13a and concave portions 13b) are generated on the surface. It is necessary to fill and crimp the melted ACF3 evenly so that there is no S gap between the terminal 11 of the printed circuit board 1 and the terminal 1 2 of the TCP2, but the remaining adhesive, interface, Due to the peeling of the film, it may not be possible to uniformly fill, and unevenness on the top surface 13 of the TCP2 is uneven at the location where such a filling failure has occurred. Therefore, the poor connection state (conducting state) between the corresponding terminals 11 and 12 becomes this unevenness and appears in the appearance.

In this embodiment, the connection portion between the TCP 2 and the printed circuit board 1 is not the connection portion between the TCP 2 and the liquid crystal glass substrate 1. Hereinafter, the inspection procedure will be described. The weight of each terminal 11 on the PCB 1 of the test body and the TCP2 electrode connected via the ACF3 The tatami area 10 is photographed by the imaging device 16 that scans. Of course, it is possible to inspect the overlapping region 10 related to the connection portion between the terminal 11 ′ of the liquid crystal substrate 1 ′ and the terminal 12 ′ of the TCP.

[0055] As shown in FIG. 3, the light irradiation to the overlapping region 10 relating to the connection portion between the TCP2 and the printed circuit board 1 is a surface light source power that irradiates light in an oblique direction with respect to the upper surface 13 of the TCP2. This is done by the illumination means 17. In this way, a wide overlapping area 10 can be dealt with. In addition, the image data 20 is acquired by the imaging device 16 on the upper surface 13 side of the TCP 2 with respect to a direction orthogonal to the surface direction of the substrate 1 and an oblique direction force that is opposite to the irradiation direction. The light irradiation direction α 1, a 1, α “... And the image acquisition direction 1) 3, β ′, / Τ... Are constant in all areas A, B, 0. (See Fig. 3.) The irradiation direction α and the acquisition direction β are oblique directions opposite to the direction perpendicular to the surface direction of the substrate 1, and both directions α and β are the surfaces of the substrate 1, respectively. The angles 0 and Θ formed with respect to the direction are not necessarily the same.

a

[0056] Since the imaging device 16 and the illuminating means 17 are integrally provided in the movable inspection head 18 as described above, the overlapping region 10 and the inspection head 18 are relatively moved, that is, the inspection head. 18 or by moving the stage S, along the scanning direction 19 (parallel direction of terminals 11 and 12 in parallel) shown in FIG.

'The image 20 for each inspection line (inspection area) can be detected in sequence with the output signal 30.

[0057] Although an area sensor is employed as the imaging device 16, in this case, it is desirable that the acquisition width of the image in the parallel direction is as narrow as possible and the acquisition pitch is as narrow as possible. . By narrowing the acquisition width, the difference in the distance between each position and the imaging device 16 in a single area where the image is acquired is reduced, and the difference in the expected angle of the camera for each position is also reduced. The resolution (inspection accuracy) of signal 30 can be improved. For this reason, the error of the output signal 30 due to the difference in the degree of coincidence of the focus within a single region can be reduced.

The thickness of the ACF 3, the width and interval of the terminals 11 on the substrate 1, the width and interval of the terminals 12 on the TCP 2, etc., differ depending on the object to be inspected. For this reason, an area sensor is used. In this case, by making the acquisition width variable, it becomes possible to deal with various types of test objects. For example, in the case of an object to be inspected with a small electrode width or electrode interval, the acquisition width can be narrowed. Conversely, in the case of an object to be inspected, the acquisition width can be increased.

[0058] By striding in this way, the overlapping region 10 is divided into a plurality of regions A, B, C-··· along the direction indicated by the arrow 19 in FIG. The region images 20a, 20b, 20c,... In the regions A, B, C ′. The region images 20a, 20b, 20c, ..., for example, as shown in FIG. 4 (a), form a set of long areas in the direction orthogonal to the scanning direction 19 indicated by the arrow, and the acquired region images 20a, The output signal 30 in the scanning direction 19 based on the image data 20 of 20b and 20 ′ ′ is obtained (see FIG. 4 (b)). The imaging device 16 obtains the image data 20 by converting it to a value that is approximately proportional to the average value of the luminance signal during scanning, so the relationship between the scanning speed and the value measured by the imaging device 16 is calculated. By doing so, it is possible to obtain from the output signal 30 the values of peak position 31, amplitude 32, period 33, and half width 34.

[0059] At this time, the region images 20a, 20b, 20c-··· are collections of data in the read direction (direction perpendicular to the scanning direction 19) and region data, respectively. In order to obtain the output signal 30 in the strike direction 19, the representative value of the data force of each area image 20a, 20b, 20c 'is calculated. In calculating the representative value, various appropriate methods can be selectively employed according to the characteristics of the object to be inspected. For example, an average value, a standard deviation value, or the like can be used.

It should be noted that “light irradiation direction α 1, α ′, 0;“ “” ”and image acquisition direction; 3, β ′, j3”... Is constantly maintained while scanning one overlapping region 10 including the amount of light irradiated to each of the regions A, B, C... And each region A, B, C-" And the area image 20a, 20b, 20c-· 'the distance to the means for acquiring', and the light receiving sensitivity of the imaging device 16 corresponding to each of the area images 20a, 20b, 20c ... The region images 20a, 20b, 20c ′... Of the regions a, b, c... Can be acquired under the same conditions.

[0060] Next, the PC. Image built in 41) or control device (controller) 40 Whether the crimping state is good or bad is determined by the processing device and the registration / determination processing device. The pass / fail judgment is performed by comparing the numerical value of the inspection item calculated by the image processing device with a pass / fail judgment reference value corresponding to the preset numerical value. From the output signal 30 of the image data 20, it is possible to determine the data such as the peak position and the width corresponding to the terminal interval by performing arithmetic processing by the image processing device. Specifically, the acquired image data 20 (a set of the region images 20a, 2Ob, 20c...) Or a set of representative values for each region 20a, 20b, 20c... Obtained from the image data 20 is scanned. Based on the output signal 30 in the direction 19, the numerical values of the peak position 31, amplitude 32, period 33, and half width 34 of the output signal 30 shown in FIG. 4 (b) can be calculated and calculated. This calculation and calculated characteristic value can be arbitrarily set in the PC 41, and calculated and calculated based on the peak position 31, amplitude 32, period 33, half width 34, and other output signals 30. A single or multiple characteristic values can be selected from the required characteristic values.

[0061] Next, characteristic values such as the peak position 31, amplitude 32, period 33, half-value width 34, etc. of the calculated and calculated output signal 30 and the calculated values based on the calculated characteristic values are the above-mentioned pair of terminals. Judgment is made based on whether or not the force exceeds the preset reference value for each of the images a, b, c.

[0062] This determination is made, for example, when the images a, b, and c shown in FIG. 4 are portions where the conduction state is good (OK portion), and the images d, e, f, and g are portions where the conduction state is bad (NG) Show)! / Images a, b, and c in the OK section are output signal 30, peak position 31, amplitude 32, period 33, half-value width 34. In the images d, e, f, and g, the characteristic values of the output signal peak position 31, amplitude 32, period 33, and half width 34 are output as irregular values.

In this way, the output signal peak position 31 in the images (areas) a, b, c ′ ″ corresponding to the terminals of the OK part and the NG part calculated from the captured area images 20a, 20b, 20c. When compared with values for amplitude 32, period 33, half-value width 34, etc., the result of the operation differs depending on the level at which any characteristic value (inspection item) force S, OK and NG can be determined. RU

[0063] Therefore, if the level deviation or all of the characteristic values become a value smaller than a predetermined threshold value, a larger value, or a value that does not fall within the upper and lower limits, the OK portion and NG The part can be clearly distinguished. · In this way, the quality of the crimped state between the corresponding terminals 11 and 12 can be judged at high speed and properly using the property that the output signal 30 represents a regular pattern.

Note that FIG. 5 is a diagram for visually explaining the difference between the image 20 in the OK portion and the image 20 in the NG portion. FIGS. 5 (a) and 5 (b) show an image 20 of the overlap region 10 including the connection portions of the terminals 11 and 12 in the OK portion, and FIGS. 5 (c) and 5 (d) show the NG portion. (b) applies image processing to (a) and (d) applies image processing to (c). Note that the reference numerals in FIG. 5 are those used for data arrangement, and are different from those described in the specification and claims of the present application.

[0065] Also, in general, the image 20 that can be captured by the imaging device 16 has a positional relationship with the unevenness generated on the upper surface 13 of the TCP2, as shown in Fig. 9 (a). The state where the specular reflection is captured at the point (the place with the steepest inclination) is brightest. On the opposite slope B point, the convex part 13a blocks the light, so it is in the darkest state.

Note that the protrusion 13a on the upper surface 13 of the TCP2 has a portion with the largest protruding height that substantially coincides with the center in the width direction (parallel direction) of the terminals 11 and 12, so that the positions of the terminals 11 and 12 are , The peak position 31 of the output signal 30 based on the image data 20 acquired from each of the regions 20a, 20b _; 20c ′. In FIG. 4 etc., the center line positions of the areas a, b, c corresponding to the terminals 11, 12 and the peak position 31 of the output signal 30 are made to coincide with each other.

'Makes the staff easy to help.

[0066] Therefore, the illumination means 17 is arranged so that the image obtained at the slope A point becomes the brightest, that is, the specular reflection light is obtained, on the object to be inspected shown in FIG. 9 (a). By adjusting the irradiation angle Θ and the acquisition angle Θ of the imaging device 16 respectively, a bright and dark image can be obtained.

However, if the irradiation angle 0 and the acquisition angle 0 set in FIG. 9 (a) are applied to other inspection objects as they are, it is not always possible to obtain a clear and bright image. If the test object is different, the thickness of each terminal 11, 12 and the width of each terminal 11, 12 and the pitch between adjacent terminals 11, 1 1, and between terminals 12, 1.2 are also different. The difference in height from 13b, pitch, slope gradient, etc.

For this reason, inspected objects with different height differences, pitches, and widths on the top surface 13 of TCP2 In the inspection, the irradiation angle Θ and the acquisition angle Θ are adjusted. For example, in the case of the specimen shown in Fig. 9 (b), it is necessary to adjust the irradiation angle to Θ 'and the acquisition angle to Θ' so that the convex slope is brightest.

In this way, the irradiation angle 0 and the acquisition angle 0 can be adjusted respectively α

Therefore, for example, it becomes possible to deal with inspected objects for various purposes such that the pitch between adjacent terminals 11 and 11 is in the range of 0.2 mm to 2. Omm.

[0068] Further, each of the area images 20a, 20b, 20c-"may be divided in a direction (read direction) orthogonal to the parallel direction! /, Respectively.

For example, when the foreign object 8 is sandwiched between the TCP2 and the substrate 1 (for example, the state shown in FIG. 11), a deformation force S such as a local bulge 13c is generated on the surface of the TCP2, so that the output signal 30 Local outlier force is also generated in the value. This abnormal value is smoothed by averaging the data within each region A, B, C-. However, if each of the regions A, B, C... Is divided into a plurality, it becomes easy to detect local fluctuations in the output signal 30.

When each region image 20a, 20b, 20c 'is divided, as shown in Fig. 10, in the divided region 20', output obtained from each region image 20a ', 20b', 20c ' Single or multiple characteristic values selected from the signal 30 'and the peak position 31', amplitude 32 ', period' 33 ', half width 34' calculated from the output signal 30 ', or the calculated value The foreign material 8 mixed between the opposing terminals 11 and 12 can be detected by the calculated value based on the characteristic value.

[0069] Further, as described above, the board 1 and the TCP 2 may be warped or bent when mounted. For example, the test object is shown in FIGS. 12 (a) and 12 (b). In this case, the states of the output signal 30 with different periods L, L · 'L, L, L are all n n + 1 m m + 1 m + 2

Can happen. Therefore, the phase shift of the output signal 30 caused by the warp or deflection is corrected.

The correction is performed based on, for example, a reference waveform obtained from a good detection target cover in which no warpage or deflection occurs in the substrate 1 or the TCP 2. The reference waveform can be calculated using theoretical calculation formulas. Terminals 11 and 12 with different characteristics are connected within a single overlapping region 10 If present, apply the reference waveform that matches terminals 11 and 12.

Based on the reference waveform, if region C corresponding to a set of terminals 11 and 12 is specified in output signal 30, region C is the starting point of output signal 30 as shown in Fig. 13 (a). ¾ point

This is the end point of the output signal 30.

However, the points E and F are both based on the reference waveform, and it can be seen that the phase of the waveform is deviated from the peak position 31, 31 "of the waveform of the output signal 30.

Therefore, the detection start point E of the output signal 30 is set at the start point (E point) based on this reference waveform. Figure 13 (b) shows the corrected waveform. Point 35 "after correction is set to point E, which is the inspection start point.

FIG. 14 (a) shows the output signal 30 in the acquired state 30, and FIG. 14 (b) shows the inspection start point E adjusted by matching the output signal 30 with the period L of the reference waveform.

0

When the period L of the reference waveform is applied to the output signal 30 adjusted for this inspection start point E, Fig. 1

0

4 As shown in (c). Therefore, the phase of each waveform Ll, L2,. '11' '' 111 appears in the output signal 30 and the phase is corrected in accordance with the period of the reference waveform (see figure).

0

Middle arrow wl, w2 '' 'w4', see).

This phase correction can be performed automatically, and by performing phase correction, even if the board 1 or TCP2 is warped or bent, the portion of the output signal 30 corresponding to each terminal 11 and 12 can be specified. Can do. For this reason, the crimping state between all the terminals 11 and 12 in the overlapping region 10 can be accurately evaluated.

FIG. 6 is a flowchart showing details of the inspection process.

[0075] First, in (Step 22), position information is extracted in advance from the PWB pattern data of the printed circuit board (board) 1 of the design drawing of the object to be inspected and used as teaching master data. A mask image is created in advance using.

In (Step 23), an image force output signal 30 is acquired by the illumination means 17 that irradiates light with an oblique force in one direction to the object to be inspected and the imaging device 16 that performs scanning.

Since the output signal 30 acquired in (Step 23) is the output signal 30 corresponding to the image for the terminal 11 of all printed circuit boards 1, it includes an area other than the overlapping area 10 with the terminal 12 of TCP2. It is out. Therefore, in (Step 24), an area corresponding to the overlapped area 10 to be inspected is specified for the output signal 30 acquired in (Step 23).

At this time, the portion that is not the overlap region 10 does not have the period 33 and amplitude 32 that are similar to the sin waveform, but appears as fine vibrations, that is, noise, so that the output corresponding to the overlap region 10 is excluded. Take signal 30 only.

Next, in (Step 25), using the reference waveform obtained in (Step 22), the phase shift of the output signal 30 obtained in (Step 23) and (Step 24) is corrected.

Next, when the overlapping area 10 is divided to inspect for contamination, etc., the process proceeds to (Step 26a) .From the output signal 30 ′ of the divided area, each terminal 11 (or TCP2) of the substrate 1 in the overlapping area 10 For each part corresponding to terminal 12), numerical values such as peak position 31 ', amplitude 32', period 33 ', half width 34', etc. are calculated and calculated. If the area is not divided, go to (Step 27a).

Next, in (Step 26b), the value of each inspection item calculated and calculated in (Step 26a) and the pass / fail judgment threshold set in advance for each inspection item, or the upper and lower inspection reference values. Make a comparison. (Step 26a) and (Step 26b) are executed for all divided regions.

[0077] Next, in (Step 27a), from the output signal 30 that was subjected to phase correction in (Step 25), it corresponds to each terminal 11 (or terminal 12 of TCP2) of the substrate 1 in the overlapping region 10 Image a, b, c-

'Every time, the peak position 31, amplitude 32, period 33, half width 34, etc. are calculated, and the numerical value is calculated.

Next, in (Step 27b), compare the value of each inspection item calculated and calculated in (Step 27a) with the pass / fail judgment threshold set in advance for that inspection item, or the upper and lower inspection reference values. I do.

[0078] Next, in (Step 28), the determination result for each inspection item compared in (Step 27b) is comprehensively determined for each of the images a, b, c. The final pass (OK) or fail (NG) is determined along with the determination result in (Step 26b). Since the peak position 31 or 31 ', amplitude 32 or 32', period 33 or 33 ', half width 34 or 34' of the NG part can be displayed easily and visually in accordance with the measurement position, It is easy to distinguish the detailed results of the NG part by numerical values. The results are immediately sent to the ACF thermocompression machine, which is the previous process. Therefore, process loss can be reduced. It should be noted that (Step 24) can be performed before (Step 23), and the order of (Step 24) and (Step 23) can be interchanged.

[0079] As described above, according to the board inspection apparatus and the inspection method of the present invention, the crimping state inspection relating to the connection portion of the terminal 12 of TCP2 connected to the terminal 11 of the printed circuit board 1 via the ACF3 is performed. Can be detected at high speed and accurately.

In addition, based on the result of inspection of the crimped state, the quality of the electrical connection state can be properly judged for all TCP2 and all terminals 11, 12; 11 ', 12' of all printed circuit boards 1. In addition, since the control device (controller) 40 or PC (personal computer) 41 that constitutes the inspection device has an arithmetic processing function, the inspection object can be obtained by performing arithmetic processing with some of the functions of the inspection device. Defects can be detected stably even when noise components due to other device elements are included.

It is possible to provide a board inspection apparatus and inspection method that can eliminate visual errors by humans, quantify pass / fail judgment criteria, and perform high-speed and accurate inspection that can detect defects with high reliability and stability. .

[0080] [Experiment 1]

FIG. 7 shows an embodiment in which the region images 20a, 20b, 20c ′... Of the regions A, B,. The gray scale in FIG. 7 (a) is an image 20 composed of a collection of area images 20a, 20b, 20c..., And the waveform in FIG.

[0081] [Experiment 2]

A method of correcting the phase shift will be described with reference to FIG. In the present invention, the amplitude of the output signal 30 is approximated by a female 1] to calculate the amplitude and other characteristic values.

[Number 1]

) '(0 = asin (iyi) where a is the amplitude and ω is the angular frequency.

However, a slight deviation may occur even when the printed circuit board 1 of the object to be inspected is aligned. In such a case, if the characteristic value is calculated based on the waveform of the output signal 30, the waveform As a result, the characteristic values such as the amplitude are not correctly calculated, and an erroneous pass / fail judgment is made. . Therefore, the waveform of the object to be inspected is detected with reference to the waveform used as a criterion for pass / fail judgment obtained by the teaching operation, and correction is performed by shifting the detected waveform by the detected phase.

[0083] If this deviation is φ, the value of ψ is obtained by the following equation.

[Equation 2] φ = tan ^-1 (CS) where C and S are as follows.

[Equation 3]

S = I y (t) sm (0t) dt

[Equation 4]

C = ^ y {t) cos ((M) dt t = 0: Start position of target signal

t = T: End position of target signal

The details of the formula [2] for detecting φ are as follows.

[0084] If the phase is shifted by φ in the expression of Woman 1], it can be expressed by the following expression.

'[Equation 5] y (t) = α η {ωΐ + φ) where> The distance between terminals is known from the reference wave, so its angular frequency ω is known. S is obtained by applying sin (cot) to woman 5] and integrating from 0 to T for one period. [Equation 6]

= acos (^) fsin ² (iyi) + asin (^) [sin (iyi) cos (iyi) <ii

Since sin (iy) cos (> t) dt = 0, substituting woman 7] into [Equation 6]

[Equation 8]

S = acos (^ ») I sin (cot) dt

, F l— cos (2 &> f)

-a cos (p) 丄 dt

=-cos () 1-^ cos (2 <yi) ^ i}

=-cos) {[t] ^T ₀ one [― sin (2 <yf)] S}

2ω

,

It becomes. Similarly, by multiplying [Equation 5] by cos (co t) and integrating from 0 to T for one period as C, and calculating in the same way as above, the following woman 9] is obtained. .

[Equation 9] αΤ

C -—— SU1 (

2 ^Υ

From [Equation 8] and [Equation 9], the formula [Equation 2] for phase φ is obtained.

Claims

The scope of the claims

[1] A plurality of terminals 11, 12; 11 ', 12' provided in parallel to the board 1 and TCP2, respectively, are overlapped, and the opposite terminals 11, 12; 11 ', 12' are anisotropic The opposing terminals 11, 12; 11 ', 12' are connected with the conductive conductor 3 interposed therebetween, and the crimping state between the opposing terminals 11, 12; 11 ', 12' is In the substrate inspection apparatus for optically inspecting the image data 20 of the overlapping region 10 of the parallel terminals 11, 12; 11 ′, 12 on the surface opposite to the side on which the terminal 12 of the TCP 2 is provided,

The overlapping region 10 is divided into a plurality of regions A, B, C... Along the parallel direction, and an oblique force is applied to each region A, B, C ′ ″ by the illumination means 17 toward the parallel direction. The image data 20 of the superimposed region 10 is acquired for each of the regions A, B, ○... By the photographing means 16 and corresponds to the regions A, B, C ′ * ′. The output signal 30 obtained from each of the region images 20a, 20b, 2 Oc "· and the characteristic value calculated from the output signal 30 or the calculated value based on the calculated characteristic value are compared with a preset reference value. The board inspection device is characterized by inspecting the crimping state between the two terminals 11, 12; 11 ', 12'.

[2] The amount of light irradiated to each of the regions A, B, C,... When acquiring the region images 20a, 20b, 20c,. The distance to the imaging means 16 and the light receiving sensitivity of the shadowing means 16, the light irradiation angle Θ and the image acquisition angle 0 with respect to the surface direction of the substrate 1 are constant in the entire areas A, B, ′. The board inspection apparatus according to claim 1, wherein:

[3] The illuminating means 17 and the photographing means 16 are integrated, and the illuminating means 17 and the photographing means 16 and the overlapping area 10 are relatively moved in the parallel direction, whereby the respective region images 20a, The board inspection apparatus according to claim 1, wherein 20b, 20c ′,... Are sequentially acquired.

[4] The characteristic value is a single characteristic value or a plurality of characteristic values selected from a peak position 31 calculated from the output signal 30, an amplitude 32, a period 33, and a half width 34, and is compared with the reference value. Is performed for each of the areas a, b, '' corresponding to the pair of terminals 11, 12; 11 ', 12' facing each other! / Board inspection device as described in any of the above.

[5] Each of the region images 20a, 20b, 20c,... Is divided in a direction orthogonal to the parallel direction, and outputs obtained from the divided region images 20a ′, 20b ′ _; 20c ′,. Single or multiple characteristic values selected from signal 30 'and peak position 31', amplitude 32 ', period 33', half width 34 'calculated from output signal 30', or calculated characteristics 4. The substrate inspection apparatus according to claim 1, wherein the foreign substance 8 mixed between the opposing terminals 11, 12; 11 ′, 12 ′ is detected by a calculation value based on the value. .

6. The phase and period of the output signals 30 and 30 ′ obtained from the superposition region 10 are corrected by comparing them with a reference signal in the same superposition region 10. The board inspection apparatus according to any one of 5 above.

[7] The imaging means 16 is an area sensor, and the acquisition width with respect to the parallel direction in each of the area images 20a, 20b, 20c- ”is variable according to the overlapping area 10 to be inspected. The board inspection apparatus according to claim 1.

[8] The substrate inspection apparatus according to any one of [1] to [7], wherein the illuminating unit 17 includes a light source having an emission spectrum substantially coincident with a light receiving sensitivity of the imaging unit 16.

[9] The irradiation angle 0 of the light by the illumination unit 17 and the image acquisition angle Θ of the image by the imaging unit 16 are respectively set in front of the TCP 2 in the overlapping region 10 to be inspected.

β

9. The substrate inspection device according to claim 1, wherein the substrate inspection device is variable in accordance with the convex shape of the surface opposite to the side on which the terminal 12 is provided.

[10] In the board inspection apparatus according to any one of claims 1 to 9, the terminals 1 and 12; 11 'and 12' of the board 1 and the TCP 2 are connected to each other through an anisotropic conductor 3. A board inspection device that is integrated with an electronic component mounting device.

[11] A plurality of terminals 11, 12; 11 ', 12, which are provided in parallel on the board 1 and TCP2, respectively, are overlapped with each other, and the opposite terminals 11, 12; 11', 12 'are anisotropic The opposing terminals 11, 12; 11 ′, 12 ′ are connected to each other with the conductive conductor 3 interposed therebetween, and the crimped state between the opposing terminals 11, 12; 11 ′, 12 ′ is In the board inspection method for optically inspecting the image data 20 of the overlapping region 10 of the parallel terminals 11 and 12; 11 'and 12' on the opposite surface of the TCP2 to the side on which the terminal 12 is provided, The overlapping region 10 is divided into a plurality of regions A, B, C... Along the parallel direction, and an oblique direction force is applied to each of the regions A, B, C. Then, the image data 20 of the superimposed region 10 is acquired for each of the regions A, B, C..., And the region images 20a, 20b, 20c ′ corresponding to the regions A, B,. By comparing the output signal 30 obtained from the above and the characteristic value calculated from the output signal 30 or the calculated value based on the calculated characteristic value with a preset reference value, the both terminals 11, 12 ; A substrate inspection method characterized by inspecting the crimping state between 11 'and 12'.

[12] The amount of light irradiated to each of the areas A, B, C ′ when acquiring the area images 20a, 20b, 20c,. The distance between the image capturing means 16 for acquiring the image data 20, the light receiving sensitivity of the image capturing means 16, the irradiation angle Θ of the light with respect to the surface direction of the substrate 1, and the image acquisition angle Θ 12. The method of inspecting a substrate according to claim 11, wherein the constant is constant at C ′.

[13] The characteristic value is a single characteristic value or a plurality of characteristic values selected from a peak position 31 calculated from the output signal 30, an amplitude 32, a period 33, and a half width 34, and is compared with the reference value. 13. The substrate inspection method according to claim 11 or 12, wherein the inspection is performed for each of the areas a, b, c ′ ′ ′ corresponding to the pair of terminals 11, 12; 11 ′, 12 ′ facing each other.

[14] Each of the region images 20a, 20b, 20c- 'is divided into directions orthogonal to the parallel direction, and obtained from the divided region images 20a', 20b ', 20c' Output signal 30 'and single or multiple characteristic values selected from peak position 31', amplitude 32 ', period 33', half width 34 'calculated from output signal 30', or calculated 13. The substrate inspection method according to claim 11, wherein the foreign matter 8 mixed between the opposing terminals 11, 12; 11 ′, 12 ′ is detected by an arithmetic value based on a characteristic value.

15. The phase and period of the output signals 30, 30 ′ obtained from the superimposition region 10 are corrected by comparing them with a reference signal in the same superposition region 10. 14! /, The board inspection method described in the gap.