JP4776197B2 - Wiring board inspection equipment - Google Patents

Description

The present invention relates to a wiring board inspection apparatus .

Japanese Patent Laid-Open No. 7-83841 Japanese Patent Laid-Open No. 11-94762 Japanese Patent Laid-Open No. 11-166903 JP-A-11-271233 JP 2002-122554 A

Semiconductor chips such as ICs and microprocessors have been rapidly integrated in recent years, so that the number of terminals at the input / output section of the chip is also increasing significantly. In response to this, electronic circuit boards for connecting such chips also have a rapidly increasing number of line conductors, and so-called build-up boards in which a plurality of wiring layers are laminated via a dielectric layer made of a polymer material. Is increasing.
In the build-up substrate, in order to electrically connect the wiring layers, a via hole penetrating the dielectric layer is formed at a necessary position of the dielectric layer, and a via conductor is formed on the inner surface of the via hole by plating.

  Conventionally, as shown in FIG. 14, this via conductor is often formed as a so-called conformal via conductor CV in which a plating layer is formed along the inner surface of the via hole and an air gap UJ is left inside thereof. In this case, when forming a structure in which another via is stacked on top of that to connect to the upper wiring layer, if the upper via conductor overlaps the gap of the lower via conductor, the plating of the via is poor. As a result, it causes a conduction failure, and therefore, the upper via conductor and the lower via conductor are formed so as not to overlap each other.

  However, as the integration density of wiring increases, the conformal via conductor as described above causes a space restriction on the upper and lower via connection structures, and recently, as shown in FIG. 15, the inside of the via conductor is filled with plating. The filled via conductor 134s has been adopted. Since the filled via conductor 134 s does not leave a gap inside the via, a so-called stacked via in which upper and lower via conductors are overlapped can be formed, and is a structure that is advantageous for high-density arrangement of vias and wirings.

  As shown in FIG. 16, the filled via conductor 134s is formed by filling the inside of the via hole with a plating metal. However, if foreign matter or the like adheres during the plating and the growth of the filled plating layer is hindered, a void remains, and the via May cause a concave defect on the top surface of the substrate. Conventionally, such a concave defect in the filled via conductor 134s has been inspected by visual observation using a magnifying glass. However, it goes without saying that efficiency and accuracy are lacking. Therefore, it is naturally conceivable to identify the concave defect by taking an image of the top surface of the via conductor and analyzing the image.

  However, since the via conductor has a metallic luster, even if it has a concave defect, when it is irradiated with illumination light, both the defect periphery and the defect inner surface generate strong reflected light, and the contrast between the defect area and the surrounding area is high. There is a difficulty that the detection accuracy of defects is lacking. In addition, Patent Documents 1 to 5 disclose various apparatuses that perform via inspection of a build-up substrate by an optical method, and all of them inspect resin residual defects inside via holes before forming via conductors. Therefore, it does not disclose an inspection method suitable for a concave defect of a filled via conductor.

The subject of this invention is providing the inspection apparatus of the wiring board which can discover the concave-shaped defect which has arisen inside the filled via conductor by plating defect etc. reliably and efficiently.

Means for Solving the Problems and Effects of the Invention

Furthermore, the inspection apparatus for a wiring board according to the present invention comprises:
A structure in which metal wiring layers and dielectric layers made of a polymer material are alternately laminated, and metal wiring layers adjacent to each other in the laminating direction with the dielectric layers interposed therebetween are connected by filled via conductors filled with metal inside An inspection apparatus for a wiring board having a wiring lamination part of
In the inspected substrate body in a state where the top surface of the filled via conductor is exposed on the surface to be inspected, the top surface region of the filled via conductor is illuminated by illumination light incident at an angle inclined from one side of the substrate main surface normal. A lighting device that illuminates;
An imaging device that captures an image of the top surface region on an imaging optical axis that is smaller than an inclined angle of the illumination light with respect to a normal surface of the substrate main surface;
A position adjusting mechanism that measures the distance to the surface to be inspected and adjusts the position of the image taking device in the direction of the optical axis of the image taking device so that the focus of the image taking device coincides with the surface to be inspected. When,
An output device for the image;
An inspection analysis unit that analyzes the occurrence state of a concave defect caused by a metal filling defect of the filled via conductor based on shadow information generated in the top surface region on the imaged image,
A plurality of the inspection substrate body a substrate workpiece which is integrated by a non-isolated form in a plane, to correct the substrate workpiece support for supporting the upper surface, the deflection of the substrate workpiece to be supported by the substrate workpiece support on A deflection correction part,
The bending correction part is a fixed holding part that holds one edge of the substrate work in a predetermined biasing direction in a plane of the substrate work arranged on the substrate work support part,
A movable holding part that is movably arranged in the biasing direction and holds the other edge in the biasing direction of the substrate workpiece;
And a tension biasing mechanism that corrects the deflection by tensioning the movable holding portion in the biasing direction to stretch the substrate work.

According to the inspection apparatus of the present invention, illumination light incident at an angle inclined from one side of the normal surface of the substrate main surface is irradiated to the filled via conductor in which the concave defect is generated. As shown in FIG. 7, the tilted illumination light causes a shadowed area by blocking the illumination light inside the recessed defect generated in the filled via conductor. Therefore, even though the inside of the concave defect is covered with a highly reflective metal, the concave defect can be easily identified by the above-described shading, and the concave defect generated inside the filled via conductor can be reliably and highly Can be discovered efficiently. Then, according to the method of manufacturing a wiring board of the present invention for selecting a substrate product by the inspection method, the detection accuracy of the concave defect is improved and the defective product can be surely eliminated. The defect rate can be reduced.

  The inspection apparatus disclosed in Patent Document 4 discloses an apparatus for inspecting a via hole using inclined illumination light. As described above, the inspection object of this apparatus is to form a via conductor by plating. This is an etching resin residue in a via hole before the process, and has a different purpose from the present invention for detecting a concave defect in a filled via conductor. In addition, since the illumination light inclined from both sides with respect to the substrate normal line is applied, when this is applied to the solution of the present invention, the shaded area caused by one inclined light is canceled by the other inclined light. Therefore, the object of detecting the concave defect by the shaded area cannot be achieved.

  The image-taking optical axis direction of the image can be tilted with respect to the normal surface of the substrate within a range smaller than the tilt angle of the illumination light, but from the viewpoint of detecting the shadow region with high accuracy, The imaging optical axis direction is preferably coincident with the normal surface of the substrate main surface.

  When the inclination angle of the illumination light with respect to the normal surface of the substrate main surface is set to 20 ° or more and 80 ° or less, the effect of improving the detection accuracy of the concave defect is enhanced. The inclination angle is more preferably set to 20 ° or more and 40 ° or less.

In the apparatus of the present invention, it is possible to use a two-dimensional image sensor such as a two-dimensional CCD sensor and to use surface illumination that illuminates the imaging region two-dimensionally. The angle of illumination light differs depending on the position, and shadow formation may be unclear, and the amount of illumination light at each position tends to be insufficient. Therefore , in the imaging area on the substrate to be inspected, select the line sensor camera as the imaging device that acquires the linear imaging information defined in the first direction in the plane, and the acquisition position of the linear imaging information Line illumination as the illumination device that illuminates with illumination light and an imaging scanning unit that scans the imaging position of the line sensor camera on the imaging area in a second direction orthogonal to the first direction in the plane And image information generating means for combining two-dimensional image information sequentially obtained by the scanning in a plane to obtain two-dimensional image information corresponding to the image area. Since the illumination light can be concentrated on the imaging region and a clearer shadow can be formed, the defect detection accuracy can be increased. In this case, when the positions of the line sensor camera and the illumination device are fixed, and the imaging scanning unit is configured to have a substrate scanning moving unit that scans and moves the substrate to be inspected in the second direction, scanning is performed from the camera optical system. This eliminates the drive mechanism for preventing the occurrence of problems such as optical axis blurring and defocusing.

In the apparatus of the present invention, by providing the position adjusting mechanism , it is possible to effectively prevent the problem that the magnification of the image changes due to the change of the focal length. In addition, the shadow area may be visually observed on the image to determine the defect, but based on the shadow information generated in the top surface area on the photographed image, the concave defect caused by the metal filling failure of the filled via conductor is detected. If an inspection analysis unit for analyzing the occurrence state is provided, the efficiency of defect determination can be further increased.

  Regarding the inspection of concave defects, specifically, the area on the image is divided into a bright area and a dark area with a predetermined threshold as a boundary by the illuminance by illumination light, and appears on the top surface area of the filled via conductor. Based on the information on the area or size of the dark region, it is possible to make a determination regarding the degree of formation of the concave defect. In this case, in the apparatus of the present invention, the inspection analysis unit described above divides the region on the image into a bright region and a dark region with a predetermined threshold as a boundary based on the illuminance by the illumination light, and the top of the filled via conductor. A dark area calculating means for calculating the area or size of the dark area appearing in the surface area and a determining means for making a determination regarding the degree of formation of the concave defect based on the area or size of the dark area are provided.

  If it carries out as mentioned above, not only the qualitative presence or absence of a concave defect but quantitative evaluation of the degree of the concave defect currently formed can be performed with the area or dimension of a dark region. As a result, it is possible to appropriately find concave defects exceeding the allowable range, and it is possible to eliminate minor concave defects that are not essentially a problem. Therefore, it is possible to appropriately reduce the defect rate in the sorting process. .

  In the case of a concave defect having substantially the same opening size, as shown in FIG. 7, since the shadow region becomes larger as the defect becomes deeper, it can be determined that the depth of the concave defect is larger as the area or size of the dark region is larger. . In the case of an apparatus, the above-described determination means is configured to determine the filled via conductor as defective when the area or size of the dark region is larger than a predetermined reference value. Since the conduction failure of the stacked via is influenced more greatly by the defect depth than the opening area of the defect, the above method that can surely eliminate the deep defect is more effective. Specifically, when the area or size of the dark region is larger than a predetermined reference value, determining the filled via conductor as defective is simple, and contributes to improving inspection efficiency.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 17 schematically shows a cross-sectional structure of a wiring board 100 that is an object of the inspection method, apparatus, and manufacturing method of the present invention. The wiring substrate 100 has a predetermined pattern on both surfaces of a plate-like core 102 made of a heat-resistant resin plate (for example, bismaleimide-triazine resin plate), a fiber reinforced resin plate (for example, glass fiber reinforced epoxy resin), or the like. Core conductor layers M1 and M11 that form wiring metal layers are formed respectively. These core conductor layers M1 and M11 are formed as a plane conductor pattern that covers most of the surface of the plate-shaped core 102, and are used as a power supply layer or a ground layer. On the other hand, a through-hole 112 drilled by a drill or the like is formed in the plate-like core 102, and a through-hole conductor 130 that connects the core conductor layers M1 and M11 to each other is formed on the inner wall surface thereof. The through hole 112 is filled with a resin hole filling material 131 such as an epoxy resin.

  In addition, wiring laminated portions L1 and L2 are formed on the upper layers of the core conductor layers M1 and M11. Specifically, first dielectric layers (build-up layers: dielectric layers) V1 and V11 each formed of a photosensitive resin composition 106 are formed, and first surfaces each having a metal wiring 107 on the surface thereof. The conductor layers M2 and M12 are formed by Cu plating. The core conductor layers M1, M11 and the first conductor layers M2, M12 are connected to each other through vias 134. Similarly, second dielectric layers (build-up layers: dielectric layers) V2 and V12 using the photosensitive resin composition 106 are formed on the first conductor layers M2 and M12, respectively. On the surface, second conductor layers M3 and M13 having metal terminal pads 108 and 118 are formed. The first conductor layers M2, M12 and the second conductor layers M3, M13 are interconnected by vias 134, respectively. As shown in FIG. 17, the via 134 is opposite to the via hole 134 h and a filled via conductor 134 s provided on the inner peripheral surface thereof, a via pad 134 p provided to be electrically connected to the filled via conductor 134 s on the bottom surface side, and the via pad 134 p. On the side, a via pad 134l projecting outward from the peripheral edge of the opening of the via conductor 34h is provided.

  The filled via conductor 134s is formed as a filled via conductor filled with a plating metal, and two or more filled via conductors 134s are overlapped with each other in a plane so as to straddle adjacent dielectric layers. Stacked vias are formed that are stacked in the resulting manner.

  On the first main surface MP1 of the plate-like core 102, the core conductor layer M1, the first dielectric layer V1, the first conductor layer M2, and the second dielectric layer V2 form the first wiring laminated portion L1. . Further, on the second main surface MP2 of the plate-like core 2, the core conductor layer M11, the first dielectric layer V11, the first conductor layer M12, and the second dielectric layer V12 form the second wiring laminated portion L2. ing. In either case, dielectric layers and conductor layers are alternately stacked such that the first main surface CP is formed of a dielectric layer, and a plurality of metals are formed on the first main surface CP. Terminal pads 110 to 117 are respectively formed. The metal terminal pad 110 on the first wiring laminated portion L1 side constitutes a solder pad that is a pad for flip-chip connecting an integrated circuit chip or the like. The metal terminal pad 117 on the second wiring laminated portion L2 side is used as a back surface pad for connecting the wiring board itself to a mother board or the like by a pin grid array (PGA) or a ball grid array (BGA). .

  The dielectric layers V1, V11, V2, and V12 and the solder resist layers 108 and 118 are manufactured as follows, for example. That is, a photosensitive adhesive film formed by forming a photosensitive resin composition varnish is laminated (bonded), and a transparent mask (for example, a glass mask) having a pattern corresponding to the via hole 134h is overlaid and exposed. The film portions other than the via hole 134h are cured by this exposure, while the via hole 134h portion remains uncured. If this is dissolved in a solvent and removed, the via hole 134h is easily formed in the intended pattern. (So-called photovia process). Next, a pattern-patterned mask is formed using a photosensitive plating resist resin, and a well-known electroless Cu plating is applied to form a wiring pattern and fill the inside of the via hole 134h with a plating metal. Thus, a filled via conductor 134s is formed. By repeating the above steps, the wiring laminated portions L1 and L2 can be formed.

  The filled via conductor 134s is formed for each dielectric layer, and each time the filled via conductor 134s is formed in one dielectric layer, it is inspected whether or not a concave defect due to poor plating filling has occurred. Therefore, the substrate body to be inspected is not the wiring substrate 100 as the final product in FIG. 17 but an intermediate product in which the process is interrupted in a state where the filled via conductor 134s is formed in the dielectric layer to be inspected. Hereinafter, the details of the inspection method according to the present invention will be described together with the apparatus configuration used therefor.

  FIG. 1 is a schematic diagram of an inspection apparatus 1 according to an embodiment of the present invention. The inspection apparatus 1 uses the illumination light (beam) LB incident at an angle inclined from one side to the main surface normal of the substrate body W to be inspected, as shown in FIG. An image of the top surface area is photographed while illuminating, and the state of occurrence of the concave defect PF caused by defective metal filling of the filled via conductor 134s is inspected based on the information of the shadow area DA generated in the top surface area on the photographed image. The illumination device 25 for generating the illumination light LB and the photographing device 13c for photographing the image of the top surface area are provided. As shown in FIG. 7, in the present embodiment, the direction of the imaging optical axis CA of the image coincides with the normal surface of the substrate main surface. As shown in FIG. 1, the inclination angle θ of the illumination light LB with respect to the normal surface of the substrate main surface is inclined in the range of 20 ° to 80 ° (desirably 20 ° to 40 °: for example, 30 °). Is set.

  The imaging device 13c includes a line sensor camera 13c that acquires linear imaging information defined in the first direction (X direction) in the plane in the imaging area on the inspected substrate W, and linear imaging information. The line illumination 25 for selectively illuminating the acquisition position of the image sensor and the imaging position by the line sensor camera 13c on the imaging area are scanned in a second direction (Y direction) orthogonal to the first direction in the plane. The imaging scanning unit 23 and an image information generation unit 12 (FIG. 3) that obtains two-dimensional image information corresponding to the imaging region by synthesizing linear imaging information sequentially obtained by the scanning in a plane. The positions of the line sensor camera 13c and the illumination device 25 are fixed, and the imaging scanning unit is configured by a substrate scanning moving unit 23 that scans and moves the inspected substrate body W in the second direction. In the present embodiment, the line sensor camera 13c is a commercially available digital line sensor camera in which a photographing optical system is added to a one-dimensional CCD sensor, and the image information generation unit 12 is also a commercially available image processing module for a digital line sensor camera (for example, AVAL DATA CORPORATION made: PSM-330D / APC332, PCI bus compatible). Further, the substrate scanning moving unit 23 is configured by a well-known XY table (hereinafter also referred to as XY table 23).

  In the present embodiment, as shown in FIG. 2, the inspected substrate W is a large-sized substrate work W in which a plurality of inspected substrates SB1, SB2,. As shown in FIG. 1, the inspection apparatus 1 includes a plurality of substrate workpiece support portions 50 and 52 that support the large-sized substrate workpiece W, and shooting positions by the shooting camera 13c. A workpiece relative movement unit 23 is provided for moving the substrate workpiece support units 50 and 52 relative to the photographing camera 13c and the illumination device 25 so as to sequentially move with respect to the inspection substrate bodies SB1, SB2,. Accordingly, the individual inspected substrate bodies SB1, SB2,... Can be sequentially inspected in the state of the large substrate work W before separation, and the trouble of mounting the inspected substrate bodies SB1, SB2,. It is efficient because it can be omitted. The work relative movement unit is also used by the XY table 23 described above.

  A large-sized substrate workpiece W in which a large number of thin inspected substrate bodies SB1, SB2,... Gather is likely to bend and easily affect defocusing of the imaging device 13c. In this case, in order to eliminate this problem, as shown in FIG. 2, a deflection correcting unit 60 that corrects the deflection of the substrate workpiece W on the substrate workpiece support units 50 and 52 is provided. The substrate work support portions 50 and 52 are both disposed on the XY table 23, and the deflection correcting portion 60 is preliminarily set in the plane of the substrate work W on the substrate work support portions 50 and 52. The fixed holding portion 50 for holding one edge of the substrate workpiece W and the substrate workpiece support portions 50 and 52 are movably disposed in the biasing direction, and the other end of the substrate workpiece W in the biasing direction is arranged. A movable holding portion 52 that holds an edge, and a stretching and biasing mechanism 53 that corrects the bending by stretching the substrate work W by pulling and biasing the movable holding portion 52 in a biasing direction. ing. By bending and stretching the ends of the substrate workpiece W on which the inspected substrate bodies SB1, SB2,... Are not formed, bending can be effectively eliminated, and the members involved in the correction are inspected substrate bodies SB1, SB2,. Since it does not interfere with the imaging area, inspection is not hindered.

  The fixed holding unit 50 is fixed on the XY table 23 by screwing or the like, and the upper surface is used as a support surface of the edge of the substrate workpiece W. On the upper surface, a positioning member inserted through a through hole SH formed in an edge region of the substrate workpiece W where the inspected substrate bodies SB1 and SB2 are not formed in order to hold and position the substrate workpiece W. A mating pin 51 is projected. On the other hand, the movable holding part 52 is slidably arranged on the XY table 23, and the same positioning engagement pin 51 is provided on the upper surface. If the through holes SH of the substrate workpiece W are inserted into the positioning engagement pins 51 and the movable holding portion 52 is urged away from the fixed holding portion 50 by the cylinder 53 forming the expansion urging mechanism, the substrate workpiece W is obtained. It is possible to add an extension tension for deflection correction.

  Returning to FIG. 1, the photographing apparatus 13c is a camera 13c whose focal length is fixedly determined in advance. Further, based on the distance measured by the distance measuring unit 21 that measures the distance of the substrate workpiece W to the surface to be inspected and the distance measured by the distance measuring unit 21, the focus of the camera 13 c matches the surface to be inspected. A camera 13c position adjustment mechanism 61 that adjusts the position of the camera 13c in the optical axis direction is provided. Accordingly, it is possible to effectively prevent the problem that the magnification of the image changes due to the change of the focal length.

  In the present embodiment, the camera 13c is disposed so as to be focused on the inspected surfaces of the substrate work support portions 50 and 52, and the illumination device 25 is inclined with respect to the photographing optical axis of the camera 13c. When the illumination beam is irradiated toward the in-focus position of the camera 13c on the surface to be inspected and the camera 13c is moved in the optical axis direction by the camera 13c position adjusting mechanism 61, the illuminating device 25 uses the camera 13c and the optical axis. It is configured to move integrally in the direction.

  The specific structure is as follows. The inspection apparatus 1 has a base 29 for installation, and a gate-shaped frame structure in which an upper frame 32 is coupled to the upper ends of columns 31, 31 erected on the base 29 is formed. Below the upper frame 32, an elevating frame 33 that can be raised and lowered along the columns 31, 31 is arranged. The elevating frame 33 is provided with female screw portions 40 and 40 so as to penetrate in the elevating direction, and the screw shafts 34 and 34 are respectively screwed together. These screw shafts 34, 34 can be rotated in conjunction with each other via a bevel gear 35 and a connecting shaft 36 in the base 29, and are rotationally driven by a camera focusing drive motor 19 provided on the upper frame 32. The elevating frame 33 is moved up and down.

  A camera 13c, a distance measuring unit 21 including a laser distance sensor, and an illumination device 25 are attached to the lifting frame 33, and these are integrated with the lifting frame 33 by driving the screw shafts 34 and 34. It is supposed to go up and down. The illuminating device 25 configured as line illumination is coupled to the camera 13c via the connecting member 38 so that the illumination beam LB faces the focal position of the camera 13c. The illuminating device 25 is disposed on a slider 39 that can be fixed on an arbitrary position around an axis orthogonal to the camera optical axis CA through the focal position on the connecting member 38, and the illumination beam LB and the camera work axis CA. The angle θ can be adjusted.

  As shown in FIG. 6, the deflection measurement reference position (for example, the center position of the workpiece) of the substrate workpiece W supported on the substrate workpiece support portions 50 and 52 is positioned immediately below the distance measurement portion 21, and up to the reference position. Measure the distance K. Prior to this measurement, the distance K0 to the horizontal reference plane P0 corresponding to zero deflection is measured in advance, and the screw shafts 34 and 34 are driven so that the focal position is located in the horizontal reference plane P0 in advance. The camera 13c is preliminarily positioned. Then, the deflection amount D is calculated by K−K0, and the screw shafts 34 and 34 can be further driven and positioned so that the focal point of the camera 13c moves downward by the deflection amount D.

  FIG. 3 is a block diagram illustrating an example of an electrical configuration of the inspection apparatus 1. The inspection apparatus 1 has an inspection analysis unit 2 that analyzes the occurrence state of a concave defect caused by a metal filling failure of a filled via conductor. The inspection analysis unit 2 includes a microcomputer (or workstation) that also serves as a drive control unit of the inspection apparatus (hereinafter also referred to as a microcomputer 2). The basic part of the microcomputer 2 includes a CPU 4, a ROM 5 that stores a computer control basic program, a RAM 6 that forms a work area of the CPU 4, and an input / output unit 7, which are connected by a host bus 3.

  Also, a PCI bus 11 is connected to the host bus 3 via a PCI bridge 10, and a card slot provided on the PCI bus 11 is connected to the substrate of the image processing module 12 that forms the image information generation unit. ing. The line sensor camera 13c is connected to the image processing module 12.

  The XY table shown in FIG. 1 has an inspection substrate W fixed on the table, which is independent of the X direction in the horizontal plane and the Y direction perpendicular to the X direction, and any XY within a specified range. It is driven so as to be positioned at the coordinate position, and includes an X drive motor 14 and a Y drive motor 18. As shown in FIG. 3, the X drive motor 14 and the Y drive motor 18 are connected to the input / output unit 7 of the microcomputer 2 via a motor driver 15 incorporating a servo control mechanism. Each of the X drive motor 14 and the Y drive motor 18 includes an encoder 16 for detecting the rotation angle position (giving the XY positioning coordinates of the inspected substrate W reflected in the motor drive amount). The motor driver 15 receives an X-coordinate command value and a Y-coordinate command value for positioning from the microcomputer 2 and compares them with the current angular positions of the motors 14 and 18 fed back from the encoder 16. It plays a role of driving the motors 14 and 18 so that the inspected substrate W moves to the target position given by the Y coordinate command value. Reference numeral 17 denotes a Schmitt trigger for sharpening the edge of the pulse signal of the encoder 16. Note that the signals of the encoders 16 of the X drive motor 14 and the Y drive motor 18 are also input to the input / output unit 7 in order to grasp the position of the inspected substrate W on the microcomputer 2 side. The above-described camera focusing drive motor 19 is connected to the input / output unit 7 together with the distance measuring unit 21 via the motor driver 20.

  A one-dimensional image signal from the line sensor camera 13c is acquired in the X direction on an imaging region (for example, a region having vertical and horizontal dimensions including only one filled via conductor) defined on the inspected substrate body W. The Then, by alternately repeating the step movement of the inspected substrate W in the Y direction and the acquisition of the one-dimensional image signal, the one-dimensional image signal can be sequentially acquired while scanning the imaging region in the Y direction. it can. The one-dimensional image signal from the line sensor camera 13c is transferred to the image processing module 12 as a series of pixel data strings, and a synchronization signal is transmitted each time transfer of one pixel data string is completed. The image processing module 12 grasps the Y-direction scanning position of the one-dimensional image signal based on the signal input from the Y-direction encoder 16 of the XY table 23 and also uses the synchronization signal from the line sensor camera 13c to determine the pixel data string. A cut is recognized, and a one-dimensional image signal at each position in the Y direction is synthesized in an image memory in the module to generate two-dimensional image data corresponding to the imaging region. The two-dimensional image data is batch transferred to the microcomputer 2 via the PCI bus 11 and stored in the image memory 6d in the RAM 6. The two-dimensional image data is stored as image data 22d in an external storage device 22 composed of an HDD or the like connected to the microcomputer 2.

  The external storage device 22 is installed with control software 22a for managing the basic operation of the inspection apparatus 1 and analysis software 22b for performing defect inspection analysis of the filled via conductor from the acquired two-dimensional image data. / Executed by the CPU 4 on the analysis software execution memory 6a.

  When inspecting the concave defect PF as shown in FIG. 7 using the inspection apparatus 1, the area on the image is illuminated by the illumination light LB, and the bright area and the dark area are set with a predetermined threshold as a boundary. And determining the degree of formation of the concave defect PF based on information on the area or size of the dark region appearing in the top surface region of the filled via conductor 134s. By executing the analysis software 22b, the inspection analysis unit 2 divides the region on the image into a bright region and a dark region with a predetermined threshold as a boundary based on the illuminance by the illumination light LB, and the filled via conductor 134s. Realizes each function of dark area calculation means for calculating the area or size of the dark area appearing on the top surface area, and determination means for determining the degree of formation of the concave defect PF based on the area or size of the dark area To do.

  The pixels included in the unitary image data from the line sensor camera 13c can be acquired as multi-bit gradation pixel data. For example, the higher the illuminance area, the greater the brightness set for the pixels in that area. If a certain threshold value is provided for the brightness setting value of a pixel, the gradation image data can be easily converted into binary image data by a known method. The bright area on the image can be specified as the first pixel aggregate area corresponding to the “bright” state in the binary image data, and the dark area can be identified as the second pixel aggregate area corresponding to the “dark” state. .

  When the degree of formation of the concave defect PF is determined by the area of the dark region, the area of the dark region can be easily calculated by counting the number of pixels forming the aggregate region of the second pixels. On the other hand, when determining by dimensions, the dimension measurement direction is determined in the second pixel assembly area, and the maximum value of the number of continuous second pixels along the dimension measurement direction is calculated, thereby determining the dimension of the dark area. Can be sought. The processing algorithm will be described later.

  In the case of the concave defect PF having substantially the same opening size, as shown in FIG. 7, since the shadow area DA increases as the defect becomes deeper, it is determined that the depth or depth of the concave defect PF increases as the area or size of the dark area increases. be able to. In this case, the dimension of the dark region can be converted into the depth H of the concave defect PF by calculation. As shown in FIG. 7, the tilted optical axis direction of the illumination light is projected onto the main surface of the substrate, the direction is set as the dimension measuring direction, and the maximum dimension L of the dark area, that is, the shadow area DA in the dimension measuring direction is set. If the inclination angle of the optical axis of the illumination light with respect to the normal surface of the substrate main surface is θ, the depth H of the concave defect PF can be estimated by Lcot θ.

  When the calculated area S or dimension L of the dark region (shadow region DA) is larger than the reference value (S0, L0), a concave defect outside the allowable range is generated in the filled via conductor 134s that acquired the dark region. This can be determined as defective. Although it is a slightly rough determination method, image binarization is performed by setting a slightly larger threshold value, and the area S or size of the dark region is dared depending on whether or not there is a pixel forming the dark region (shadow region DA). It is also possible to perform defect determination without calculating L. In this case, a good (pass) determination is made only when all of the pixels are in the bright area, so the product of the pixel set values in the defined area of the image is calculated and the determination is made. become.

  Next, in the inspection apparatus 1 in FIG. 3, a determination result storage unit 22 that stores the formation position information of the filled via conductor on the inspected substrate body and the determination result of the quality of the filled via conductor in association with each other, and is set on the output region. Determination result output units 8 and 9 for mapping and outputting the pass / fail determination result are provided at positions corresponding to the filled via conductors 134s in the substrate region thus formed. In this way, it is possible to grasp at a glance at which position of the inspected substrate bodies SB1, SB2,... The filled via conductor 134s determined to be defective. The formation position information of the filled via conductor on the substrate to be inspected is stored in the external storage device 22 as product-specific via position data 22c. As shown in FIG. 4, the data 22c is prepared for each product number (product type) of the wiring board, and in order to grasp each via position different for each dielectric layer, each filled via conductor for each layer of the dielectric layer. Coordinate data (for example, center position coordinates of a circular via region).

  The determination result storage unit is formed as a storage unit for the via defect determination result data 22e in the external storage device 22. As shown in FIG. 5, the inspection determination result of the filled via conductor includes the product number of the wiring board, the lot number (lot specifying information) of the large substrate work W, the substrate number (substrate specifying information) in the large substrate work W, and the filled via conductor. The defect area, the defect depth, and the pass / fail judgment result of each filled via conductor are stored in association with the layer number (layer identification information) of the dielectric layer in which is located. In addition, photographed image data (reference numeral 22d in FIG. 3) of each filled via conductor is also stored.

  The coordinate data of the filled via conductor of the designated layer is read from the via position data 22c by product, and the pass / fail judgment result of each filled via conductor is read from the via defect judgment result data 22e (determination result storage unit 22). As shown in the figure, the determination result of each filled via conductor at the corresponding coordinate position in the output area is a figure that can distinguish the pass / fail result from each other (for example, the color and shape of the plot points are changed between the pass / fail judgment VA and the fault judgment VR). ) To map. The mapping result can be output from a monitor 8 or a printer 9 as a determination result output unit connected to the input / output unit 7 in FIG.

  Hereinafter, the flow of the image acquisition process of the inspection apparatus 1 will be described using a flowchart. FIG. 9 shows the flow of the image acquisition process, which is executed by the control software 22a of FIG. First, when the large-sized substrate workpiece W is set on the substrate-work support portions 50 and 52 of the XY table 23 and the processing is started, the deflection correcting unit 60 is operated first, and the large-sized substrate workpiece W is expanded to correct the deflection. (See FIG. 2). Next, S1: Initial setting processing in FIG. 9 starts.

  FIG. 10 shows the details. In step S51, the part number and lot number of the substrate are input. In step S52, the number of the layer to be inspected is input. Read from the via position data 22c (FIG. 4). Next, in S53, the XY table 23 is driven to move the large-sized substrate workpiece W to the origin position. In S54, the camera position adjustment mechanism 61 is operated, and the focus of the camera 13c is set to the origin position (horizontal reference in FIG. Positioning in the height direction is performed so as to coincide with the plane P0). In step S56, the distance measurement unit 21 measures the distance L to the large-sized substrate workpiece W, and calculates the camera movement amount L-L0 (described above) to be adjusted in step S57. In step S58, the camera moves by the calculated movement amount. 13c is lowered to focus on the surface of the substrate workpiece W.

  The initial setting process is thus completed, and the process proceeds to S2 in FIG. 9 to initialize the number of the inspected substrate. In S3, the large substrate work W is moved by the XY table 23, and the in-substrate origin position (for example, an alignment mark formed on the substrate) of the inspection target substrate body of the designated number in the large substrate work W is obtained. 3), and the X counter 6b and Y counter 6c in FIG. 3 are reset in S4.

  In S5, the via number is initialized. In S6, the camera 13c is aligned with the filled via conductor of the designated number, and in S7, the field of view is set so that the entire filled via conductor is inside. Then, an image of the filled via conductor is captured by the method already described in S8, and the captured image is stored in S9. In other words, instead of taking an image of the entire surface of the substrate to be inspected, for each position where the filled via conductor is formed in the substrate, a limited field of view in which the filled via conductor is accommodated is set and image photographing processing is performed individually. To do. As a result, it is possible to eliminate a useless image area in which no filled via conductor exists, and even if the resolution of a captured image of the filled via conductor is increased, the image size does not have to be enlarged.

  Next, in S10, it is determined whether or not there is a next via, and if there is, the process proceeds to S11, the number of the via is incremented, the process returns to S6, and the processes up to S10 are repeated. If there is no next via in S10, the imaging process of the inspected substrate body is terminated, and the process proceeds to S12 to check whether or not an uninspected inspected substrate body remains on the large-sized substrate workpiece W. If it remains, it progresses to S13 and increments the number of a to-be-inspected board | substrate body, returns to S3, and repeats the process to S12 below. If the next substrate to be inspected does not remain in S12, the image acquisition process is terminated.

  FIG. 11 shows the flow of a quality determination process for a filled via conductor using the acquired image. First, in S101, identification information (article number, lot number, substrate number, and layer) of a substrate to be inspected as a determination target is input, and image data of the filled via conductor of the corresponding layer is read in S102. In S103, the via number is initialized, and the image of the filled via conductor having the number designated in S104 is binarized. In S105, determination processing is performed using the binarized image data.

  FIG. 12 shows the flow of processing when the determination processing is performed based on the area of the shadow region. First, in S201, on the binarized image, the setting state of the pixels in the shadow area is set to “1”, and the setting state of the pixels outside the shadow area is set to “0” (dark outside the outline of the filled via conductor) If there is a background area, an edge line between the background area and the filled via conductor area is extracted by a known image processing method, and only the inside of the edge line is set as a determination target area). ”Is obtained, and this is used as the area parameter S of the shadow region. Next, in S202, the area parameter S is compared with a reference value S0 for pass / fail determination. If S ≦ S0, the filled via conductor is determined to be acceptable, and if S> S0, it is determined to be unacceptable. Stored as shown in FIG. 5 (S203 to S205).

  On the other hand, FIG. 13 shows the flow of processing when the defect depth H is obtained from the shaded area and the determination is made based on the defect depth H. Mainly the process of determining the dimension measurement direction already described with reference to FIG. 7 as the X direction and obtaining the continuous number of pixels that are shadow areas (that is, “1”) for each pixel (pixel) column in the X direction. Become. Specifically, the Y-direction position is initialized in S251, and in S252, the continuous number L ′ of pixels forming the shaded area of the X-direction pixel row having the designated Y coordinate is obtained, and the Y-direction position is sequentially determined in S253. The process of S252 is repeated while incrementing. When the above processing is completed for all the pixel columns in the image area, the process proceeds from S254 to S255, where the maximum value of the continuous number L ′ of pixels forming the shadow area is obtained and specified as the dimension L of the shadow area in the dimension measurement direction. To do. In S256, the defect depth H is calculated by H = Lcot θ, and in S257, this is compared with the reference value H0. If H ≦ H0, the filled via conductor is determined to be acceptable, and if H> H0, it is determined to be unacceptable, and the result is stored as shown in FIG. 5 (S258 to S260). Since cot θ is a constant when the illumination angle θ is constant, the process of comparing the shadow area dimension L itself with the reference value L0 is mathematically equivalent, and the determination can be made using this.

  Via failure determination result data 22e tabulated as shown in FIG. 5 is output from the monitor 8 or printer 9 shown in FIG. As shown in S107 to S108 in FIG. 11, the position of each via is read out and the determination result is plotted at the position corresponding to the via in the mapping area, whereby the output result as shown in FIG. 8 is obtained. Then, referring to the output result, for each lot of substrates to be inspected, that is, for each large-sized substrate workpiece, a mark indicating that the substrates to be inspected are defective is indicated by marking with ink or the like. Form. After inspection, the large-sized substrate work is cut and separated into individual wiring boards, and the substrate to be inspected without defective marks can be obtained by manually observing the inspected substrate body with defective marks or using a well-known sorting device. Can be sorted and separated from the body.

Hereinafter, experimental results performed to confirm the validity of the inspection method of the present invention will be described. Prepare about 10000 intermediate products of various wiring boards with filled vias with a height of 35-40 μm and a diameter of about 80 μm, and specify the filled vias with a concave defect with a depth of 15 μm or more in advance using an optical microscope. Oita. Next, when the filled via having a concave defect of 15 μm or more is specified while variously changing the line illumination angle θ by the inspection apparatus of FIG. The test judgment was repeated with the case of successful and unmatched as unsuccessful. The specifications of each part of the inspection device are as follows.
(Line sensor camera)
Number of pixels: 4096 pixels / line Resolution: 5 μm / pixel Focal length: 125 mm
(Line lighting)
-Uses halogen light and a lens with a focusing distance of 100 mm.
The results are shown in Table 1.

  According to the above results, the probability of successful inspection is high when the illumination angle θ is 20 ° or more and 80 ° or less, and the success probability is even higher when the angle θ is 20 ° or more and 60 ° or less. It can be seen that a success probability of almost 100% is obtained at 20 ° or more and 60 ° or less.

The front schematic diagram which shows one Embodiment of the test | inspection apparatus of this invention. The schematic diagram which shows the state which set the large-sized board | substrate workpiece | work on XY table shape. The block diagram which shows an example of an electrical structure of the test | inspection apparatus of FIG. Conceptual diagram of via position data by product. The conceptual diagram of via defect determination result data. The measurement conceptual diagram of the board | substrate bending amount. The principle explanatory drawing of the inspection method of the present invention. The conceptual diagram which maps a via inspection result. The flowchart which shows the flow of an image acquisition process. 10 is a flowchart showing details of the initial setting process of FIG. 9. The flowchart which shows the flow of the quality determination process of a via. 12 is a flowchart showing a first example of details of the determination process of FIG. 11. The flowchart which shows the 2nd example of the detail of the determination process of FIG. Conceptual diagram of conformal beer. Conceptual diagram of filled via conductor and stacked via using the same. The schematic diagram which shows the generation | occurrence | production condition of the concave defect to a filled via conductor. The schematic diagram which shows the cross-section of the wiring board used as the application object of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Inspection apparatus 12 Image processing unit (Image information generation means)
13c Line sensor camera (photographing device)
22 External storage device (judgment result storage unit)
25 Illumination Device 100 Wiring Board L1, L2 Wiring Laminating Part 134s Filled Via Conductor SB1, SB2 ... Substrate Body LB Illumination Light PF Concave Defect DA Shaded Area W Large Size Board Work

Claims (6)

A structure in which metal wiring layers and dielectric layers made of a polymer material are alternately laminated, and metal wiring layers adjacent to each other in the laminating direction with the dielectric layers interposed therebetween are connected by filled via conductors filled with metal inside An inspection apparatus for a wiring board having a wiring lamination part of
In the inspected substrate body in which the top surface of the filled via conductor is exposed on the surface to be inspected, the top surface region of the filled via conductor is illuminated with illumination light incident at an angle inclined from one side of the substrate main surface normal line. A lighting device;
An imaging device that captures an image of the top surface region on an imaging optical axis that is smaller than an inclined angle of the illumination light with respect to a normal surface of the substrate main surface;
A position adjusting mechanism that measures the distance to the surface to be inspected and adjusts the position of the image taking device in the direction of the optical axis of the image taking device so that the focus of the image taking device coincides with the surface to be inspected. When,
An output device for the image;
An inspection analysis unit that analyzes the occurrence state of a concave defect caused by a metal filling defect of the filled via conductor based on shadow information generated in the top surface region on the imaged image,
A plurality of the inspection substrate body a substrate workpiece which is integrated by a non-isolated form in a plane, to correct the substrate workpiece support for supporting the upper surface, the deflection of the substrate workpiece to be supported by the substrate workpiece support on A deflection correction part,
The bending correction part is a fixed holding part that holds one edge of the substrate work in a predetermined biasing direction in a plane of the substrate work arranged on the substrate work support part,
A movable holding part that is movably arranged in the biasing direction and holds the other edge in the biasing direction of the substrate workpiece;
An inspection apparatus for a wiring board, comprising: an extension biasing mechanism that corrects bending by pulling and biasing the movable holding portion in the biasing direction to stretch the substrate work.   A line sensor camera as the imaging device that acquires linear imaging information defined in a first direction in a plane in an imaging area on the substrate to be inspected, and an acquisition position of the linear imaging information Line illumination as the illumination device that selectively illuminates with illumination light, and a photographing position by the line sensor camera on the photographing region in a second direction orthogonal to the first direction in the plane An imaging information scanning unit that scans and image information generation means that obtains two-dimensional image information corresponding to the imaging region by combining the linear imaging information sequentially obtained by the scanning in a plane. The wiring board inspection apparatus according to 1. The inspection analysis unit divides the region on the image into a bright region and a dark region with a predetermined threshold as a boundary according to the illuminance by the illumination light, and appears on the top surface region of the filled via conductor. Dark area calculation means for calculating the area or size of the dark area,
The wiring board inspection apparatus according to claim 1, further comprising: a determination unit configured to determine a degree of formation of the concave defect based on an area or a dimension of the dark region.   The wiring board inspection apparatus according to claim 3, wherein the determination unit determines that the filled via conductor is defective when the area or size of the dark region is larger than a predetermined reference value. A determination result storage unit that stores the formation position information of the filled via conductor on the inspected substrate body and the quality determination result of the filled via conductor in association with each other;
The wiring board inspection apparatus according to claim 4, further comprising: a determination result output unit that maps and outputs the pass / fail determination result at a position corresponding to each filled via conductor in a substrate region set on the output region.   Work relative movement for relatively moving the substrate work support relative to the image capture device and the illumination device so that the photographing position by the photographing device sequentially moves with respect to a plurality of substrate bodies to be inspected on the substrate work support. The wiring board inspection apparatus according to claim 1, further comprising: a section.

Images