JP2008203144A - Gas sensor manufacturing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a gas sensor manufacturing method capable of suppressing axial deviations which occur between a main body metal fixture and a detection element due to defective solid powder materials interposed between the main body metal fixture and the detection element, when gas sensors are manufactured. <P>SOLUTION: A solid powder material, supplied in a solid state in a manufacturing process of gas sensors, is arranged as a sample on a detection base (S13) and photographed hierarchically as a camera is made to approach by one unit stepwise (S16-S20). Contrast is compared between the same pixels in images of each hierarchy, and hierarchies that are in focus on the pixels are specified, to specify pixels focusing on chip parts which can occur in the solid powder materials. The size (volume value) of chips is determined, on the basis of the pixels (S21) and is compared with a volume reference value (S27, S28) to identify solid powder materials, in which chips of sizes that are not in conformity with specifications have occurred and the solid powder materials can be removed from a manufacturing line of gas sensors (S31). <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a method of manufacturing a gas sensor in which a detection element for detecting the concentration of a specific gas in exhaust gas is held inside a metal shell.

  Conventionally, a gas sensor provided with a detection element that generates electromotive force of different magnitude or changes its resistance value according to the concentration of a specific gas, such as NOx (nitrogen oxide) or oxygen, in an exhaust gas of an automobile or the like It has been known. For example, a detection element using zirconia has a solid electrolyte body made of zirconia in a bottomed cylindrical shape, and the oxygen concentration is based on the difference in oxygen partial pressure between the air introduced inside and the exhaust gas filled outside. The detection is performed. Such a detection element is held by a metal shell and is attached to the exhaust pipe of an automobile. Since the attachment is generally performed by screwing into the mounting hole of the exhaust pipe, the metal sensor metal fitting is attached to the outer peripheral surface. It is formed in the cylinder shape which has a screw thread.

  In recent years, in order to reduce the size and performance of gas sensors, plate-like detection elements have been developed instead of bottomed cylindrical detection elements. In such a gas sensor using a cylindrical detection element and a plate detection element, in order to prevent the exhaust gas from escaping to the outside through the inside of the metal shell, the gap between them is filled with a powder material made of talc powder. ing. In the manufacturing process of the gas sensor, in order to facilitate the handling of the powder material made of such powder, a solid powder material is produced by pressing and compacting the powder material in a cylindrical shape in advance. And when this gas sensor is assembled, this solid powder material is arranged, crushed, pulverized, filled into details, and compressed to enhance the airtightness through the metal shell (see, for example, Patent Document 1).

Such a solid powder material is made into a cylindrical shape by pressing it into a substantially uniform density using a mold. Then, the solid powder material is put into a component supply device called a feeder, and at the time of manufacturing the gas sensor, one solid powder material is taken out from the feeder and supplied to the production line for manufacturing one gas sensor.
JP 2001-281209 A

  However, in the process of manufacturing the gas sensor, the solid powder material is put into the feeder accommodating portion together with the other solid powder material, and each solid powder is taken out from the feeder and supplied. Since the powder materials come into contact with each other, the edge portions and the like are easily worn. The progress of the wear may cause chipping of the solid powder material, and in particular, if there is a biased chipping centering around the insertion hole of the detection element, the chipping will not occur on the side where the chipping occurs when assembling to the metal shell. The amount of powder material is reduced compared to the non-generated side. As a result, when the solid powder material is compressed and deformed, the detection element receives stress toward the uneven chipped side, and the axis of the metal shell and the axis of the detection element may be displaced.

  If the gas sensor is assembled with such an axial misalignment, the separator for taking out the electrode of the detection element during assembly may place a load on the detection element, and the detection element may be lost or broken. there were.

  The present invention has been made in order to solve the above-described problems. When the gas sensor is manufactured, the metal shell and the detection element are separated by a defect in the solid powder material interposed between the metal shell and the detection element. It aims at providing the manufacturing method of the gas sensor which can suppress the axial shift which may arise in the meantime.

  In order to achieve the above object, a method of manufacturing a gas sensor according to a first aspect of the present invention is a rod-shaped or cylindrical shape that extends in the axial direction and detects a specific component in a gas to be detected at its tip side. A detection element having a detection portion, a cylindrical metal shell surrounding the detection element, and a powder material disposed between the outer peripheral surface of the detection element and the inner peripheral surface of the metal shell. A gas sensor manufacturing method comprising: supplying the powder material as a cylindrical solid powder material in which powder is preliminarily compacted while having an insertion hole through which the detection element is inserted in the manufacturing process of the gas sensor. A first photographing step in which photographing of a first surface which is a surface on one opening end side of the insertion hole of the solid powder material is performed, and the other opening of the insertion hole of the solid powder material Shooting of the second surface, which is the end surface Image analysis processing is performed on the captured image of the first surface of the solid powder material imaged in the second imaging step and the first imaging step, and a chip generated on the first surface side of the solid powder material A first image analysis step in which the size of the solid powder material is detected, and an image analysis process is performed on the photographed image of the second surface of the solid powder material photographed in the second photographing step, and the solid powder material A second image analysis step in which the size of a chip generated on the second surface side is detected, and a first volume value corresponding to the size of the chip on the first surface side of the solid powder material is calculated. A first volume value calculating step, a second volume value calculating step for calculating a second volume value corresponding to the size of the chip on the second surface side of the solid powder material, the first volume value and the second volume value. When both volume values are smaller than a predetermined volume reference value, Serial solid powder material is characterized in that a step conformance determination is determined to be within specification.

  According to a second aspect of the present invention, in addition to the configuration of the first aspect of the invention, the detection element has a plate shape having a rectangular cross section perpendicular to the axial direction, and the solid powder. The opening shape of the insertion hole opened in the first surface and the second surface of the material forms a rectangular shape in accordance with the shape of the detection element. In the first volume value calculating step, The first surface of the powder material was cut in the longitudinal direction of the opening shape of the insertion hole and divided in the short direction, and each half surface corresponded to the size of the chip contained in the half surface. Two of the first volume values are calculated, and in the second volume value calculation step, the second surface of the solid powder material is cut in the longitudinal direction of the opening shape of the insertion hole, and is divided in the short direction, For each half plane, the size of the chip included in the half plane Two corresponding second volume values are calculated, and in the standard conformity determining step, all of the two first volume values and the two second volume values are smaller than a predetermined half volume reference value. The solid powder material is determined to be within specifications.

  In the manufacturing process of the gas sensor, the powder material used when holding the detection element in the metal shell is supplied as a solid powder material. In order to detect a chip that may occur in an edge portion or the like in the supply process, the powder material is detected. As in the gas sensor manufacturing method according to the present invention, each of the first surface and the second surface of the solid powder material is photographed, and image analysis processing is performed on the photographed image, so that the size of the chip generated in the solid powder material is reduced. Detection can be performed. If, for example, a known contrast method, lens focus method, confocal method, or the like is used as the image analysis processing method, the size of a chip generated in the solid powder material can be detected from a captured image of the solid powder material. Further, the first volume value corresponding to the size of each chip on the first surface side and the second surface side in a non-contact state with respect to the solid powder material from the size of the chip generated in the detected solid powder material, It is possible to determine the second volume value. Then, if the obtained first and second volume values are respectively compared with the volume reference values, and only the solid powder material for which the size of the generated chip is determined to conform to the standard is used for the production of the gas sensor, it is substantially omitted in the metal shell. The powder material can be filled with a uniform packing density. As a result, it is possible to suppress misalignment between the two when the detection element is held in the metal shell, and it is possible to improve the yield of the gas sensor.

  By the way, when the detection element has a plate shape, the rigidity when external stress is applied from the short direction in the cross section in the axial direction is lower than the rigidity when external stress is applied from the longitudinal direction. When the powder material is arranged in the cylindrical metal shell, if there is a difference in the packing density of the powder material in the short direction in the cross section in the axial direction of the detection element, there is a difference in the packing density in the longitudinal direction Compared to the above, the detection element is likely to be bent or broken. Therefore, as in the invention according to claim 2, the photographed image of the solid powder material is cut in the longitudinal direction of the opening shape of the insertion hole in the cross section of the detection element, and is divided for each half surface divided into two in the short direction. 1 and 2nd volume value is calculated | required and each is compared with a half volume reference value. Thereby, when a chip occurs in the solid powder material, the first and second volume values when the chip is formed on the short side of the cross section of the detection element are both the same volume chip. It takes a larger value than the first and second volume values when it is formed across the half surface. Therefore, when the half volume reference value is used as a threshold value and compared with the first and second volume values, even if the solid powder material has the same volume chip, it is biased toward the short side of the cross section of the detection element. If a chip is formed and a gas sensor assembly failure is likely to occur, it is judged that it does not meet the standard. If a chip is formed on the longitudinal side of the cross section of the detection element and it is difficult to cause a gas sensor assembly failure, it conforms to the standard. If it judges that it will, it can improve the yield of a gas sensor.

  Hereinafter, an embodiment of a gas sensor manufacturing method embodying the present invention will be described with reference to the drawings. First, as an example of a gas sensor manufactured by the method for manufacturing a gas sensor according to the present invention, the structure of a gas sensor 1 will be described with reference to FIG. FIG. 1 is a longitudinal sectional view of the gas sensor 1. The gas sensor 1 is used by being attached to an exhaust pipe (not shown) of an automobile. At this time, the side exposed in the exhaust pipe (the lower side in FIG. 1) is the tip side in the direction of the axis O, and the opposite side. Assume that (the upper side in FIG. 1) is the rear end side.

  A gas sensor 1 shown in FIG. 1 is attached to an exhaust pipe (not shown) of an automobile, and a detection unit 11 provided on the front end side of a detection element 10 held inside is exposed to exhaust gas flowing in the exhaust pipe. The so-called full-range air-fuel ratio sensor detects the air-fuel ratio of the exhaust gas from the oxygen concentration in the exhaust gas. When the air-fuel ratio of the exhaust gas is lean, a detection value (current value) corresponding to the amount of oxygen surplus with respect to the theoretical air-fuel ratio is obtained from the detection element 10, and when it is rich, unburned gas A detection value (current value) corresponding to the amount of oxygen necessary to completely burn the gas is obtained. Based on these detection values, the air-fuel ratio of the exhaust gas is obtained by a sensor control circuit (not shown) and output to an ECU (electronic control unit) for use in air-fuel ratio feedback control and the like.

  First, the detection element 10 will be described. The detection element 10 is a narrow and plate-like element extending in the direction of the axis O as is well known, and includes a gas detection body that detects the oxygen concentration, and a heater that heats the gas detection body for early activation. The body is integrated as a laminated body bonded in the thickness direction (in FIG. 1, the left-right direction on the paper surface is shown as the thickness direction (plate thickness direction), and the front-back direction on the paper surface is shown as the width direction). The gas detector is composed of a solid electrolyte body mainly composed of zirconia and a detection electrode mainly composed of platinum (both not shown), and the detection electrode is disposed in the detection section 11 of the detection element 10. And in order to protect a detection electrode from poisoning by exhaust gas, the protective layer 15 is formed in the detection part 11 of the detection element 10 so that the outer peripheral surface may be wrapped. Further, five electrode pads 16 for taking out electrodes from the gas detection body and the heater body are formed at the rear end portion 12 of the detection element 10. In the present embodiment, the detection element 10 is described as a “detection element” in the present invention. However, strictly speaking, a heater body is not necessarily required as the configuration of the detection element 10, and a gas detection body is not necessarily included in the present invention. It may correspond to a “detection element”.

  Next, the flange portion 24 will be described. A metal cup 20 made of metal having a bottomed cylindrical shape with the detection element 10 inserted therein is disposed slightly on the front end side of the central portion 13 of the detection element 10. The metal cup 20 is a holding member for holding the detection element 10 in the metal shell 50, and the detection portion 11 of the detection element 10 protrudes from the opening 25 at the bottom of the cylinder. Moreover, the front-end | tip peripheral part 23 of the edge part of a cylinder bottom is formed in the taper shape over the outer peripheral surface. A ceramic holder 21 made of alumina and a sealing material 22 made of talc powder are accommodated in the metal cup 20 with the detection element 10 inserted therethrough. The sealing material 22 is crushed in the metal cup 20 to be filled in detail, whereby the metal cup 20, the ceramic holder 21 and the sealing material 22 are integrated, and the flange portion 24 serves as the radial direction of the detection element 10. It is assembled together in a form that surrounds it.

  Next, the metal shell 50 will be described. The metal shell 50 is for attaching and fixing the gas sensor 1 to an exhaust pipe (not shown) of the automobile, and has a cylindrical shape in which a through hole 58 is formed. The center portion 13 of the detection element 10 is held in the through hole 58 of the metal shell 50 together with the flange portion 24. The metal shell 50 is made of, for example, low carbon steel such as SUS430, and a male screw portion 51 for attachment to the exhaust pipe is formed on the outer peripheral tip side. A distal end engaging portion 56 to which a protector 8 described later is engaged is formed on the distal end side of the male screw portion 51. In addition, a tool engaging portion 52 that engages a tool for attachment is formed at the center of the outer periphery of the metal shell 50, and between the front end surface of the tool engaging portion 52 and the rear end of the male screw portion 51. A gasket 55 for preventing gas escape when attached to the exhaust pipe is inserted. Further, the rear end side of the tool engaging portion 52 is engaged with a rear end engaging portion 57 to be engaged with an outer cylinder 65 described later, and the detection element 10 is caulked and held in the metal shell 50 on the rear end side. For this purpose, a caulking portion 53 is formed.

  Further, a stepped portion 54 having a step shape is formed in the vicinity of the male screw portion 51 on the inner periphery of the through hole 58 of the metal shell 50. The step 54 is engaged with the peripheral edge 23 of the metal cup 20 that forms the flange 24 integrated with the detection element 10. Further, a sealing material 26 made of talc powder is loaded on the inner periphery of the metal shell 50 from the rear end side of the flange portion 24 in a state where the detection element 10 is inserted through the metal shell 50 itself. A cylindrical sleeve 27 is fitted into the metal shell 50 so as to hold the sealing material 26 from the rear end side. A shoulder portion 28 having a step shape is formed on the outer periphery of the rear end side of the sleeve 27, and an annular caulking packing 29 is disposed on the shoulder portion 28. In this state, the crimping portion 53 of the metal shell 50 is crimped so as to press the shoulder portion 28 of the sleeve 27 toward the distal end side via the crimping packing 29. The sealing material 26 is crushed in the metal shell 50 and filled in details, and the flange portion 24 and the detection element 10 are mainly composed of the sealing material 26 and the sealing material 22 loaded in advance in the metal cup 20. It is positioned and held in the metal fitting 50. The sealing materials 22 and 26 correspond to the “powder material” in the present invention.

  Next, the protector 8 will be described. As described above, the detection portion 11 of the detection element 10 held inside protrudes from the tip (tip engagement portion 56) of the metal shell 50. The tip engaging portion 56 protects the detecting portion 11 of the detecting element 10 from contamination caused by deposits (toxic substances such as fuel ash and oil components) in exhaust gas, breakage due to water, and the like. A protector 8 is fitted and fixed by spot welding or laser welding. The protector 8 is a double structure constituted by a bottomed cylindrical inner protector 90 and a cylindrical outer protector 80 that surrounds the periphery in the radial direction with a gap between the outer peripheral surface of the inner protector 90. Have

  The inner protector 90 includes a plurality of inner introduction holes 95 on the rear end side of the peripheral wall 92, a plurality of drain holes 96 on the front end side of the peripheral wall 92, and a discharge port 97 on the bottom wall 93 (wall portion on the front end side). Is open. The base end portion 91 on the opening end side (rear end side) is engaged with the outer periphery of the distal end engaging portion 56 of the metal shell 50, and laser welding is performed around the outer periphery simultaneously with the outer protector 80 described later. The inner protector 90 is fixed to the metal shell 50.

  Further, the outer protector 80 has a plurality of outer introduction holes 85 opened on the distal end side of the peripheral wall 82. The base end portion 81 on the opening end side is engaged with the outer periphery of the base end portion 91 of the inner protector 90, and laser welding is performed simultaneously with the inner protector 90. The outer protector 80 is also connected via the inner protector 90. It is fixed to the metal shell 50. Furthermore, the tip 83 of the outer protector 80 is bent inward toward the peripheral wall 92 of the inner protector 90 so as to close the gap between the outer protector 80 and the inner protector 90.

  The gap between the outer protector 80 and the inner protector 90 causes the exhaust gas introduced from the outside through the outer introduction hole 85 to generate a swirling flow in a state of surrounding the outer periphery of the peripheral wall 92 of the inner protector 90, and thereby the gas component It is provided to separate water and moisture. The gas component is introduced into the inner protector 90 from the inner introduction hole 95, contacts the detection element 10, and is discharged to the outside from the discharge port 97, while moisture enters the inner protector 90 from the drain hole 96. , And is configured to be discharged from the discharge port 97 to the outside. With this configuration, the detection unit 11 of the detection element 10 is protected from contamination due to deposits in the exhaust gas, breakage due to thermal shock caused by moisture, and the like.

  Next, the structure of the rear end side of the metal shell 50 of the gas sensor 1 will be described. From the rear end (caulking portion 53) of the metal shell 50, the rear end portion 12 of the detection element 10 held inside protrudes. The rear end portion 12 is covered with a cylindrical separator 60 made of insulating ceramics. The separator 60 holds therein a plurality of connection terminals 61 that are brought into contact (electrically connected) with the plurality of electrode pads 16 formed on the rear end portion 12 of the detection element 10. Each connection terminal 61, a plurality of lead wires 64 connected to each connection terminal 61 and drawn out of the gas sensor 1 and each connection portion are also accommodated and protected.

  The outer cylinder 65 is made of stainless steel (for example, SUS304) and has a cylindrical shape. The outer cylinder 65 is attached to the rear end side of the metallic shell 50 and is exposed from the rear end portion 12 of the detecting element 10 and the separator 60 exposed from the rear end of the metallic shell 50. It covers and protects. The outer cylinder 65 has an opening end 66 on its front end side engaged with the outer periphery of the rear end engaging portion 57 of the metal shell 50 and is crimped from the outer peripheral side, and makes a round around the outer periphery, and the rear end engaging portion 57. And is fixed to the metal shell 50 by laser welding.

  Further, a metal-made cylindrical holding metal fitting 70 is disposed in the gap between the outer cylinder 65 and the separator 60. The holding metal fitting 70 has a support portion 71 formed by bending the rear end of the holding member 70 inward, and a support portion 71 is provided with a flange portion 62 provided in a hook shape on the outer periphery of the rear end side of the separator 60 inserted into the holding metal fitting 70. And the separator 60 is supported. In this state, the outer peripheral surface of the outer cylinder 65 at the portion where the holding metal fitting 70 is disposed is crimped, and the holding metal fitting 70 that supports the separator 60 is fixed to the outer cylinder 65.

  A fluorine rubber grommet 75 is fitted into the opening on the rear end side of the outer cylinder 65. The grommet 75 has five insertion holes 76 (one of which is shown in FIG. 1), and the five lead wires 64 drawn from the separator 60 are inserted into each insertion hole 76 in an airtight manner. ing. In this state, the grommet 75 is crimped from the outer periphery of the outer cylinder 65 while pressing the separator 60 toward the front end side, and is fixed to the rear end of the outer cylinder 65.

  Next, a manufacturing process of the gas sensor 1 will be described with reference to FIGS. FIG. 2 is a diagram showing the configuration of the computer 230 and its peripheral devices. FIG. 3 is a diagram illustrating a manufacturing process of the gas sensor 1. FIG. 4 is a diagram illustrating a manufacturing process of the gas sensor 1. FIG. 5 is a diagram illustrating a manufacturing process of the gas sensor 1. FIG. 6 is a flowchart of the main routine of the standard conformity judgment program. FIG. 7 is a flowchart of image analysis processing of the standard conformity determination program. FIG. 8 is a diagram for explaining the first and second imaging steps.

  FIG. 9 is a diagram illustrating an image virtually formed by image data obtained by binarizing a sample captured image at the focus position A. FIG. FIG. 10 is a diagram illustrating an image that is virtually formed by image data obtained by binarizing a captured image of a sample at the focus position B. FIG. FIG. 11 is a diagram showing an image virtually formed by image data obtained by binarizing a sample photographed image at the focus position C. FIG. FIG. 12 is a diagram showing an image virtually formed by image data obtained by binarizing a sample captured image at the focus position D. FIG. FIG. 13 is a diagram showing a silhouette image of a sample virtually formed by binarized image data. FIG. 14 is a diagram showing an image virtually formed by image data obtained by binarizing the shape of the missing portion of the sample at the focal position B. FIG. FIG. 15 is a diagram showing an image virtually formed by image data obtained by binarizing the shape of the missing portion of the sample at the focal position C. FIG. FIG. 16 is a diagram showing an image virtually formed by image data obtained by binarizing the shape of the missing portion of the sample at the focus position D. FIG.

  In the manufacturing process of the gas sensor 1, the sealing materials 22 and 26 are supplied as a solid sealing material 260 (see FIG. 3) that is pressed and consolidated into a ring shape. The solid sealing material 260 may be chipped when supplied to the production line (not shown) of the gas sensor 1, and in this embodiment, it is inspected whether the solid sealing material 260 conforms to the standard. This inspection is performed in accordance with a standard conformity determination program (see FIG. 6) executed by a CPU 231 mounted on a known computer 230 (see FIG. 2). Details of the inspection by the standard conformity determination program will be described later, but prior to that, the configuration of the computer 230 on which the CPU 231 that executes the standard conformity determination program and its peripheral devices will be described. The solid sealing material 260 corresponds to the “solid powder material” in the present invention.

  As shown in FIG. 2, the computer 230 includes a known CPU 231 that controls various controls, a ROM 232 that stores a program such as BIOS executed by the CPU 231, a RAM 233 that temporarily stores data when the CPU 231 processes data, and The computer is a known computer that is connected to an I / O bus 248 that mediates data transfer and configured to execute various programs. Connected to the I / O bus 248 are a keyboard 246 and a mouse / pointing device 247 for inputting operations, and a hard disk drive (hereinafter referred to as “HDD”) 235 which is a data storage device.

  The I / O bus 248 is also connected to other devices such as a monitor drive device for displaying a screen and a drive device for reading and writing data storage media such as a DVD-ROM. Further, an input / output port 249 is also connected to the I / O bus 248, and signals are transmitted / received to / from an external device via the input / output port 249, and when the solid sealing material 260 is inspected. Control of operations of various devices to be driven is performed. Specifically, a camera driving device 221, an inspection table arranging device 222, a reverse arranging device 223, and a nonconforming product sorting device 224 are connected to control the operation timing of each device. The camera driving device 221 is a device for moving the CCD camera 220 for photographing the solid sealing material 260 up and down in the Z-axis direction. The inspection table arranging device 222 removes the solid sealing material 260 at the time of photographing. It is a device for placing on top. Further, the reverse placement device 223 is a device that reverses the solid sealing material 260 and rearranges the solid sealing material 260 for photographing the bottom surface 266 side after photographing the top surface 265 side (see FIG. 3) of the solid sealing material 260. The product sorting device 224 is a device for removing from the production line the solid sealing material 260 that is determined not to conform to the standard in a standard conformity determination program described later. Furthermore, image data captured by the CCD camera 220 is also taken into the computer 230 via the input / output port 249 and stored in the captured image data storage area 236 of the HDD 235.

  Here, a standard conformity judgment program is stored in the program and initial setting value storage area 242 of the HDD 235 described above, and various storage areas used by the standard conformity judgment program are secured when the standard conformity judgment program is executed. Is done. Specifically, the captured image data storage area 236, the binarized image data storage area 237, the contour image data storage area 238, the deletion part image data storage area 239, the first volume value storage area 240, and the second surface volume value. A storage area 241 is secured. The HDD 235 is also provided with various storage areas not shown.

  In the photographed image data storage area 236, bitmap data of a photographed image on the top surface 265 side or the bottom surface 266 side (see FIG. 4) of the solid sealing material 260 photographed by the CCD camera 220 is stored. As will be described later, this photographed image is photographed while slightly shifting the in-focus position. In this embodiment, four photographed images having focal positions A, B, C, and D (see FIG. 8) are used. An image storage area is provided. In the binarized image data storage area 237, image data binarized from the image data stored in the photographed image data storage area 236 by what is in focus for each pixel and what does not match is stored. The Four image data storage areas respectively corresponding to images taken at the four focal positions A to D are provided.

  In the contour image data storage area 238, the image data stored in the binarized image data storage area 237 is overlaid, and the silhouette image data (2 of the solid sealant 260 formed by synthesizing the focused pixels). (Valued image data) is stored. The deletion part image data storage area 239 is obtained based on each image data stored in the photographed image data storage area 236 and the image data of the silhouette of the solid sealing material 260 stored in the contour image data storage area 238. Image data (binarized image data) indicating the shape of a chipped portion (deleted portion) that can occur in the solid sealing material 260 is stored. As described above, there are provided storage areas for image data of four deletion portions respectively corresponding to images photographed at four focus positions A to D. The first volume value storage area 240 and the second surface volume value storage area 241 are generated on the top surface 265 side of the solid sealing material 260 calculated based on each image data stored in the deletion portion image data storage area 239. The chip volume to be obtained and the chip volume to be generated on the bottom surface 266 side are stored.

  Further, in the program and the initial set value storage area 242 described above, a unit volume value (described later) used in a standard conformity determination program described later, and used to obtain a missing volume that may occur in the solid sealing material 260, A volume reference value as a reference value to be compared with the calculated volume of the missing portion is stored. In a predetermined storage area of the RAM 233, an inversion flag used for confirming which side of the sample, the top surface 265 side or the bottom surface 266 side, is photographed at the time of photographing is stored.

  In the manufacturing process of the gas sensor 1, the solid seal material 260 is separated by the computer 230 and the peripheral devices configured as described above, and only the standards-compliant product is used for assembling the gas sensor 1. Hereinafter, the details of the manufacturing process of the gas sensor 1 will be described. In the following, a process in which the sealing material 22 and the sealing material 26 constituting the gas sensor 1 are supplied at the time of assembling the gas sensor 1 will be described in detail, and manufacturing processes of other parts of the gas sensor 1 are well known and will not be described. Alternatively, the description will be simplified.

  In the manufacturing process of the gas sensor 1, as shown in FIG. 1, the detection element 10, the ceramic holder 21, and the sealing material 22 are inserted into the metal cup 20 to produce the flange portion 24. Thereafter, the sealing material 26 and the sleeve 27 are inserted from the rear end side of the detection element 10 to which the flange portion 24 is assembled, and in this state, the seal member 26 and the sleeve 27 are disposed in the through hole 58. The packing 29 is disposed between the detection element 10 and the metal shell 50 from the rear end 12 side, and the crimping portion 53 is crimped, whereby the detection element 10 is held in the metal shell 50. As described above, the sealing material 26 is made of talc powder, but in the process of manufacturing the gas sensor 1, the solid sealing material 260 (see FIG. 3) that is pressed into a ring-like cylinder is easy to handle. ).

  As shown in FIG. 3, by passing the talc powder into the mold 200 and compressing it, an insertion hole 261 through which the detection element 10 is inserted is formed between the cylindrical top surface 265 and the bottom surface 266. A cylindrical solid sealing material 260 is formed (solid powder material forming step). The formed solid sealing material 260 is put together with other solid sealing materials 260 formed in the same manner into a housing portion of a supply device (hereinafter referred to as “feeder” 210) to a production line (not shown). In the feeder 210, an arbitrary one is taken out from the plurality of solid sealing materials 260 accommodated in the accommodating portion, and supplied to the production line (solid powder material supplying step).

  Since a plurality of the solid sealing materials 260 are put into the accommodating portion of the feeder 210, the solid sealing materials 260 may collide with each other in the accommodating portion, thereby causing wear. In particular, a corner portion (a ridge corner portion formed by the outer peripheral surface, the top surface, and the bottom surface) is likely to be cut, and a defect may occur in part around the corner portion. Therefore, in the manufacturing process of the gas sensor 1, a non-defective inspection is performed on the solid sealing material 260 taken out from the feeder 210. As shown in FIG. 4, the solid sealing material 260 taken out from the feeder 210 is transported onto an inspection table (not shown) but provided as a specific part on the supply line. 2 (see FIG. 2), the top surface 265, which is the surface on one opening end side of the insertion hole 261, is positioned upward and positioned at a predetermined position on the inspection table (upward in the Z-axis direction in FIG. 4). Note that the top surface 265 and the bottom surface 266 of the solid sealing material 260 are distinguished for convenience in order to specify a surface to be imaged in a first imaging process and a second imaging process, which will be described later. Either of the surfaces may be the top surface 265.

  Incidentally, a CCD camera 220 for photographing the solid sealing material 260 disposed on the inspection table is provided above the inspection table in the Z-axis direction so as to be movable in the Z-axis direction. The CCD camera 220 has a fixed focal length F (see FIG. 8) in advance, and the imageable range (size of the XY plane orthogonal to the Z-axis direction) is constant. An image photographed by the CCD camera 220 is reproduced by tens to millions of pixels arranged at equal intervals, and the area of the photographing surface represented by one pixel is photographed at a focal length F. It is obtained in advance from the relationship between the size of the range to be captured and the actual size value of the shooting range.

  Using such a CCD camera 220, photographing is performed on the top surface 265 side of the solid sealing material 260 disposed on the inspection table (first photographing step). Shooting is performed according to the processing of a standard conformity determination program described later, but the CCD camera 220 is moved from the initial position downward (in the direction approaching the inspection table) by a certain distance (hereinafter, the movement distance is referred to as “1 unit”). Whenever one unit moves, the top surface 265 side of the solid sealing material 260 on the examination table is photographed. The captured image of the solid sealing material 260 is transmitted to the computer 230.

  Then, when the photographing of the top surface 265 side of the solid sealing material 260 is finished, the reverse placement device 223 (FIG. 2) is arranged so that the bottom surface 266 of the solid sealing material 260 faces upward (in the Z-axis direction) according to the processing of the standard conformity determination program. The top and bottom of the solid sealing material 260 are reversed by the above (referred), and are positioned and arranged at the predetermined position on the inspection table (reversing step).

  Similarly to the first photographing process, the CCD camera 220 performs photographing on the bottom surface 266 side of the solid sealing material 260 (second photographing process). That is, according to the processing of the standard conformity determination program described later, the CCD camera 220 is moved one unit at a time from the initial position downward (in the direction approaching the inspection table), and each time the bottom surface 266 side of the solid sealing material 260 is photographed. Is called. The captured image of the solid sealing material 260 is transmitted to the computer 230.

  In the standard conformity determination program, image analysis processing is performed on the image of the top surface 265 side and the bottom surface 266 side of the photographed solid sealing material 260 to detect the size of a chip that may occur in the solid sealing material 260, Based on this size, the volume of the chipping (deletion part) that can occur in the solid sealing material 260 is calculated. Although this process will be described later, the standard conformity determination program determines whether or not the solid sealing material 260 conforms to the standard based on the calculated missing volume. Then, as shown in FIG. 5, the solid sealing material 260 that does not conform to the standard is removed from the production line of the gas sensor 1 (standard conformity determination step). On the other hand, the solid sealing material 260 conforming to the standard is conveyed on the production line, and in the next step, it is fitted as the sealing material 22 into the metal cup 20 in which the ceramic holder 21 is inserted, or the flange portion 24 is assembled thereafter. The detection element 10 is fitted as a sealing material 26 from the rear end portion 12 side (assembly process). Thereafter, as shown in FIG. 1, the intermediate assembly 2 in which the sleeve 27 is fitted is disposed in the metal shell 50 to which the protector 8 is joined, the packing 29 is assembled, and the crimping portion 53 is crimped. Thus, the solid sealing material 260 is crushed and filled in detail as powder sealing materials 22 and 26, and the detection element 10 is held by the metal shell 50.

  In the subsequent manufacturing process of the gas sensor 1, the outer cylinder 65 with the separator 60 and the grommet 75 being caulked and held therein is assembled therein, and the open end 66 of the outer cylinder 65 goes around its outer periphery to engage the rear end of the metal shell 50. The gas sensor 1 is completed by laser welding to the portion 57.

  As described above, in the processes from the first imaging process to the standard conformity determination process in the manufacturing process of the gas sensor 1 described above, each process is performed according to the execution of the standard conformity determination program. The details of each process of the standard conformity judgment program of FIGS. 6 and 7 will be described below with reference to FIGS. 2 to 5 and FIGS. 9 to 15 as appropriate.

  The standard conformity determination program shown in FIG. 6 is stored in the HDD 235 program and initial setting value storage area 242 of the computer 230 shown in FIG. 2, and is read into the RAM 233 by the CPU 231 and executed. It is executed when the manufacture of the gas sensor 1 is started, and an initialization process is first performed (S10). In this initialization process, storage areas such as various flags and variables used in the standard conformity determination program are secured in a predetermined storage area of the RAM 233, and each storage area (captured image data storage area) is stored in the HDD 235. 236, a binarized image data storage area 237, a contour image data storage area 238, a deletion part image data storage area 239, a first volume value storage area 240, and a second surface volume value storage area 241) are secured. In addition, a reset signal is sent to each device (camera drive device 221, inspection table placement device 222, reversal placement device 223, and nonconforming product sorting device 224), and control of each device is started.

  Next, the process waits until the supply of the solid sealing material 260 from the feeder 210 to a production line (not shown) is detected by, for example, a microswitch disposed on the production line (S11: NO). When the supply is detected (S11: YES), the first volume value storage area 240 and the second surface volume value storage area 241 secured in the HDD 235 are initialized and the reverse flag is reset (S12). Next, a control signal is transmitted to the inspection table arranging device 222 so that the supplied solid sealing material 260 is arranged as a sample to be inspected at a predetermined position on the inspection table (not shown) (S13). Further, initialization of image data and image data (specifically, a photographed image data storage area 236 secured in the HDD 235, a binary image data storage area 237, a contour image data storage area 238, and a deletion portion image data storage) Area 239 is initialized) (S15). In the following, a sample having a chip shape such as the solid sealing material 260 shown in FIG. 8 will be described as an example.

  Next, a control signal is transmitted to the camera driving device 221 (see FIG. 2), and the CCD camera 220 is moved to the initial position (S16). As described above, the CCD camera 220 shown in FIG. 8 is adjusted so that the focal length becomes F. The initial position of the CCD camera 220 is that the focal position A is at least a sample (solid seal) in the Z-axis direction. It is determined to be a position of the surface (top surface 265) of the material 260) or a position higher than that in the Z-axis direction (a direction away from the inspection table). At this position, the CCD camera 220 captures the sample on the top surface 265 side, and the image data of the captured image (digital image) is “focused on position A in the captured image data storage area 236 (see FIG. 2). The image data "is stored in a storage area reserved for saving (S17).

  As described above, the sample is shot while moving the CCD camera 220 by one unit downward in the Z-axis direction (direction approaching the inspection table), but the shooting is continued until it is moved to the position determined as the end of shooting. (S19: NO). For this reason, a control signal is transmitted to the camera driving device 221 and the CCD camera 220 is moved one unit downward in the Z-axis direction (S20). The focus position of the CCD camera 220 is B. Then, the process returns to S17, and the sample is photographed on the top surface 265 side, and the photographed image data is stored in the storage area of “image data focusing on position B” in the photographed image data storage area 236 (see FIG. 2). (S17).

  Thereafter, the same processing is repeated, and the image data captured by moving the focus position of the CCD camera 220 to C is stored in the storage area of “image data focused on position C” in the captured image data storage area 236 ( S19: NO, S20, S17). Then, the image data captured with the focus position moved to D is stored in the storage area of “image data with focus on position D” in the captured image data storage area 236 (S19: NO, S20, S17). . In this embodiment, the case where photographing is performed with the focal point positions A to D being taken as an example, the CCD camera 220 is moved and the position of the CCD camera 220 when the focal point position becomes D is shown. It is defined as a shooting end position. Accordingly, when the photographing of the sample at the focal position D is completed, it is determined that the photographing is finished (S19: YES), and a subroutine of the image analysis process is called (S21).

  As shown in FIG. 7, in the subroutine of the image analysis process, first, a process of specifying the in-focus position of the sample is performed (S41). The photographed image on the top surface 265 side of the sample (solid sealing material 260) photographed while shifting one unit at a time in the Z-axis direction has a hierarchical structure in which the sample is photographed in layers with the focus positions A, B, C, and D. It can be said to be an image. Therefore, pixels having the same XY coordinates are extracted from the image data of each layer composed of a plurality of pixels. Next, the contrast value of each extracted pixel is compared, and the pixel which shows a peak value is specified. Further, the contrast value of the specified pixel is compared with a predetermined threshold value, and if the value is greater than that, it is determined that the photographed image is in focus in the hierarchy including the specified pixel in the XY coordinates. To do. The binarized image data storage area 237 is provided with a focus flag corresponding to each pixel of each layer, and the focus flag corresponding to the specified pixel is set to 1 in the XY coordinates where the focus position is specified. The

  When this process is performed on all the pixels while shifting the XY coordinates, 2 is obtained from the image data of the photographed image according to the focus flag value (1 or 0) as shown as a virtual image in FIGS. Binarized image data of a sample having a stratified value is obtained. As shown in FIG. 8, the focus position A is in a position above the surface of the sample (top surface 265) in the Z-axis direction, and there is no focused pixel in the image obtained by photographing this. Accordingly, the binarized image data constituted by the focus flag value stored in the “binarized image data focusing on position A” in the binarized image data storage area 237 is the virtual image shown in FIG. As shown in the image, all the focus flags are shown as 0 images. The focus position B is at the position of the surface of the sample in the Z-axis direction, and is in focus except for pixels corresponding to portions where chipping may have occurred. Accordingly, the binarized image data constituted by the value of the focus flag stored in the “binarized image data focusing on the position B” in the binarized image data storage area 237 is the virtual image shown in FIG. As shown in the image, only the focus flag of the pixel corresponding to the surface (the top surface 265) excluding the portion where the chipping could occur is shown as an image of 1.

  Next, the focal point position C is in a position corresponding to the approximate center of a chipped portion that may occur on the top surface 265 side of the sample in the Z-axis direction, and a part of the pixel corresponding to the chipped portion is focused. Therefore, the binarized image data constituted by the value of the focus flag stored in the “binarized image data focusing on the position C” in the binarized image data storage area 237 is the virtual image shown in FIG. As shown in the image, the image is shown as an image having pixels with the focus flag set to 1 in a part of the pixels corresponding to the portion where the chipping may have occurred. The focal point position D is a portion of a chip that may occur on the top surface 265 side of the sample in the Z-axis direction and is located on the side far from the top surface 265, and the other part of the pixel (focal point) corresponding to the portion where the chipping may have occurred. The portion in focus C at the position C) is in focus. Accordingly, the binarized image data constituted by the value of the focus flag stored in the “binarized image data focusing on the position D” in the binarized image data storage area 237 is the virtual image shown in FIG. As shown in the image, it is shown as an image having a pixel whose focus flag is 1 in the other part of the pixel corresponding to the portion where the chipping may have occurred.

  Note that the contrast value of the specified pixel is compared with a predetermined threshold value because the background pixel (the pixel that is not in focus at any level in the Z-axis direction) included in the image of the sample is focused. This is to prevent a flag having a value of 1 from being included.

  Next, as shown in FIG. 7, a process of forming a silhouette shape of the sample (solid sealing material 260) is performed by superimposing the binarized image data of each layer (S42). By removing pixels having a contrast value less than a predetermined threshold when obtaining sample binary image data as described above, that is, pixels corresponding to the background and not in focus at any level, each level is excluded. The virtual image obtained by superimposing the virtual images of the binarized image data is an image showing the silhouette of the sample viewed from the Z-axis direction as shown in FIG. In this process, for the binarized image data stored in the binarized image data storage area 237, the logical sum (OR) of the focus flags of the pixels having the same XY coordinates is obtained, and the obtained contour is obtained. Image data is stored in the contour image data storage area 238.

  Next, as shown in FIG. 7, a process of specifying the position of the surface (top surface 265) of the sample (solid sealing material 260) is performed (S43). In this process, with respect to the binarized image data stored in the binarized image data storage area 237, the presence / absence of a pixel whose focus flag is 1 in the shooting order (position A, B, C, D order). Is confirmed. First, if “binarized image data focusing on the position A” is confirmed, and there is no pixel whose focus flag is 1, it is determined that the focal position A is above the position of the surface of the sample in the Z-axis direction. Is done. Then, “binarized image data focusing on position B” in the binarized image data storage area 237 is similarly confirmed. In the present embodiment, the case where the position of the top surface 265 of the sample is at the focus position B is an example, and therefore there is a pixel with a focus flag of 1, and in the processing of S43, the position B is the position of the surface of the sample. Identified.

  In the next S44, a process of specifying the shape of the portion where the sample may be missing for each layer is performed (S44). In this process, the binarized image data of each layer stored in the binarized image data storage area 237 and the contour image data stored in the contour image data storage area 238 are focused on the pixels having the same XY coordinates. An exclusive OR (XOR) of the flags is determined. First, the binarized image data having the focus on the position (position B) of the surface of the sample specified in S43 is compared with the contour image data obtained in S42. Then, the exclusive OR of the focus flags of the pixels having the same XY coordinates is obtained and stored in the corresponding storage area of the deletion portion image data storage area 239 (here, “deletion portion image data focusing on the position B”). Remembered. As shown in the virtual image of FIG. 14, this “deleted part image data with the focus on position B” is obtained by projecting a portion where the chip of the sample may have occurred onto the XY plane at position B. The image is composed of pixels corresponding to the portion where the chipping may have occurred.

  Further, this “deleted portion image data with focus at position B” is compared with the next “binarized image data with focus at position C” in the order of shooting, and the pixels having the same XY coordinates as described above. Is obtained and stored in a corresponding storage area of the deletion part image data storage area 239 (here, “deletion part image data with focus on position C”). As shown in the virtual image of FIG. 15, this “deleted part image data with the position C as a focus” is obtained by projecting a portion where the chip of the sample may have occurred on the XY plane at the position C. The image is composed of pixels corresponding to the portion where the chipping may have occurred.

  Similarly, “position D obtained by calculating an exclusive OR of“ deleted portion image data with focus on position C ”and the next“ binarized image data with focus on position D ”in the order of photographing. The deletion portion image data having a focus on “is also stored in the deletion portion image data storage area 239. In the present embodiment, a virtual image (see FIG. 15) composed of “deletion part image data with focus on position C” and a virtual image composed of “binarized image data with focus on position D”. Since the images (see FIG. 12) coincide with each other, a virtual image constituted by “deleted portion image data with focus at position D” obtained by obtaining an exclusive OR of both is as shown in FIG. The image is composed of pixels having a focus flag of 0. In S44, when the missing portion in each layer is projected onto the XY plane in this way, the missing portion image data that can be constituted by pixels corresponding to the portion where the lack may occur is created. Thereafter, the process returns to the main routine. It should be noted that the CPU 231 executes the process of S44, so that the process of creating the deletion portion image data corresponding to the missing portion of the sample is the “first image analysis step” and “second image analysis step” in the present invention. Is equivalent to.

  Returning to the main routine of the standard conformity judgment program of FIG. 6, the process proceeds to S22, and using the deleted portion image data obtained in this way, the area of the missing portion of each layer is obtained, and in the layer direction (Z-axis direction). A process of obtaining the chipped volume by adding together is performed (S22). As described above, the area of the photographing surface represented by each pixel constituting the image photographed by the CCD camera 220 is obtained in advance. In addition, if the moving distance per unit of the CCD camera 220 moved at the time of photographing is also obtained in advance, by multiplying both (area and moving distance), the missing portion expressed by one of the focus flags can be obtained. A three-dimensional size, that is, a volume (or a volume value corresponding to the size) is obtained. The volume of the missing portion that can be expressed by one focus flag is obtained in advance as a unit volume value, and the value of the focus flag of the deletion portion image data in each layer stored in the deletion portion image data storage area 239 is determined. By multiplying all the sums, in S22, a process of calculating the volume (first volume value) of the chipped portion that can occur on the top surface 265 side of the sample is performed. In addition, the process of calculating | requiring the volume (volume value) of the missing part which may arise in a sample by the process of S22 performed by CPU231 is called "the 1st volume value calculation process" and the "2nd volume value calculation process" in this invention. Equivalent to.

  Next, in S23, it is confirmed whether or not the value of the inversion flag stored in the predetermined storage area of the RAM 233 is 1 (S23). Since 0 is stored in the inversion flag in S12 (S23: NO), the process proceeds to S24 and 1 is stored in the inversion flag (S24), and a control signal is transmitted to the inversion placement device 223 to invert the sample. And rearrangement is performed (S25). The top surface 265 and the bottom surface 266 are reversed in the Z-axis direction by the reverse placement device 223, and the solid sealing material 260 is placed again at a predetermined position on the inspection table (not shown).

  Then, returning to S16, this time, photographing of the bottom surface 266 side of the solid sealing material 260 (S15 to S20), and calculation of the volume of the chipped portion (second volume value) that may have occurred on the bottom surface 266 side by image analysis processing. Performed (S21). Since the inversion flag at this time is 1 (S23: YES), the process proceeds to S27.

  First, a first volume value indicating the volume of a chipped portion that may have occurred on the top surface 265 side of the sample (solid sealing material 260) calculated through image analysis processing and stored in the first volume value storage area 240, and a program The volume reference value stored in advance in the initial set value storage area 242 is compared (S27). The volume reference value prevents the shaft of the metal shell 50 and the axis of the detection element 10 from being displaced due to the chipping that may occur in the solid sealing material 260 when the gas sensor 1 is assembled, and prevents assembly failure. The allowable range of the size (volume) of the chip is determined. When the first volume value is equal to or greater than the volume reference value (S27: NO), it is determined that the sample is a nonconforming product, and a control signal is transmitted to the nonconforming product sorting device 224 to remove the sample from the production line. Is performed (S31). Even if the first volume value is less than the volume reference value (S27: YES), the volume of the chipped portion that may have occurred on the bottom surface 266 side of the sample stored in the second volume value storage area 240 in S28. If the 2 volume value is equal to or greater than the volume reference value (S28: NO), similarly, a process of removing this sample from the production line (not shown) is performed (S31). If both the first volume value and the second volume value are less than the volume reference value (S27: YES, S28: YES), it is determined that the sample is a conforming product, and a control signal is transmitted to the nonconforming product sorting device 224. And the process which returns this sample to a production line is performed (S29). After the process of S29 or S31, the process returns to S11, waits for the solid sealing material 260 as the next sample to be supplied from the feeder 210 (S11: NO), and if supplied (S11: YES), the above steps are performed. The solid sealing material 260 is inspected repeatedly.

  The present invention is not limited to the above-described embodiments, and various modifications can be made. In this embodiment, as an example of detection of missing samples by image analysis processing, a contrast method is performed from image data of the solid sealing material 260 photographed hierarchically by the CCD camera 220 in accordance with a standard conformity determination program executed by the CPU 231 of the computer 230. In this way, the in-focus pixel is determined to detect the chipping, but the present invention is not limited to this method. For example, the solid sealing material 260 is photographed hierarchically using an analog camera, the photographed image is partitioned with a mesh filter, passed through a high-pass filter, and a known lens focus method is used to find out the focused section, thereby detecting chipping. May be performed. Alternatively, a pinhole array in which a plurality of pinholes provided with microlenses are arranged is attached to an analog camera, and the focal position is obtained from the light intensity when the reflected light irradiated to the solid seal material 260 passes through the pinhole array. The confocal method may be used to determine the in-focus position and detect the chipping. Alternatively, the surface of the solid sealing material 260 may be scanned with laser light, and the surface shape of the chipped portion may be determined from the phase difference of the reflected light to detect the chipped portion. Other known image analysis methods may be used to detect chipping.

  Further, in the present embodiment, the sample is photographed in the four layers, but the photographing may be performed by further dividing into more layers. In this case, the focused position may move back and forth in the Z direction depending on the unevenness on the surface of the sample. Therefore, prior to the processing of S43 of the image analysis processing, the number of pixels whose focus flag is 1 is confirmed in the order of photographing the binarized image data of each layer stored in the binarized image data storage area 237. Then, a layer in which the focus flag of pixels more than a predetermined number of pixels is 1 is specified as a layer corresponding to the position on the surface of the sample (for example, position B), and binarized image data of the layer is designated as the layer. It is overwritten by the logical sum with the binarized image data of the previously confirmed hierarchy (for example, position A). Furthermore, if the binarized image data of the previous layer (upper layer) specified as the position of the sample surface is initialized and then the processing of S43 is performed, the irregularities on the sample surface are erroneously recognized as missing. Therefore, it is possible to more accurately determine the chipping.

  Further, in the present embodiment, as the first volume value calculating step, the area per pixel obtained from the actual size value of the photographing surface by the CCD camera 220 and the number of pixels constituting the photographed image, and the CCD camera 220 at the time of photographing. From the moving distance, the volume of the missing portion of the sample expressed by one pixel was obtained. And the standard conformity judgment of the sample was performed by comparing the volume and a volume reference value. In this way, the volume of the missing portion of the sample itself may be obtained. However, the number of pixels representing the missing portion of the sample, that is, the deleted portion image of each layer stored in the deleted portion image data storage area 239. You may add the value of the focus flag which comprises data, and make it the volume value corresponding to a volume, and you may compare this with a volume reference value (what converted the volume of the chip | tip which becomes a threshold value into the number of pixels).

  Further, based on the opening shape of the insertion hole 261 (see FIG. 8) of the solid sealing material 260, the sample image data and the binarized image data are divided into two in the short direction, and the volume calculated for each half plane. By comparing the value with the volume reference value, it may be determined whether or not the solid sealing material 260 is within the standard. Specifically, when the binarized image data indicating the silhouette of the sample is created in S42 of the image analysis process, the pixel portion corresponding to the insertion hole 261 of the solid sealing material 260 is recognized using a known image recognition method. I do. Then, the longitudinal direction and the short direction are detected from the shape constituted by the pixels corresponding to the insertion hole 261, and the binarized image data is divided so that the silhouette shape of the sample is bisected in the short direction. . For example, FIG. 17 shows a virtual image obtained by dividing the image formed from the pixels constituting “binary image data with the focus on position B” shown in FIG. 10 into two parts as described above. In the virtual image S on one side and the virtual image T on the other side divided in the short direction (Y direction) of the shape of the insertion hole 261, the chipping generated in the sample is biased toward the virtual image S side. In this modification, the volume value of the chipped portion on the virtual image S side and the volume value of the chipped portion on the virtual image T side are obtained by image analysis processing similar to that of the present embodiment, respectively. Compared with the volume reference value, if the standard is not satisfied, the sample is judged as a nonconforming product.

  As described above, whether the sample image data or the binarized image data is divided into two halves is determined according to the standard because the detection element 10 of the gas sensor 1 has a plate shape and the axis line. This is because the cross-sectional shape in the O direction is rectangular, and the rigidity when external stress is applied in the direction orthogonal to the axis O varies depending on the direction of the external stress. For example, in the virtual image S and the virtual image T similar to FIG. 17, as shown in FIG. 18, the vicinity of one end side in the longitudinal direction (X direction in the figure) of the shape of the sample insertion hole 261 (see FIG. 8) is the center. When chipping occurs, in the gas sensor manufactured using this sample, the packing density of the sealing material (see FIG. 1) is low in the vicinity of the chipped portion. In the detection element of this gas sensor, external stress is applied in the longitudinal direction in a cross section orthogonal to its own axis O. On the other hand, in a gas sensor manufactured using a sample having a chip at the position shown in the virtual image of FIG. 17, external stress is applied to the detection element in the lateral direction in a cross section orthogonal to its own axis O. .

  By the way, when both have the same chip volume, the external stress applied to the detection element is the same. However, in view of the structure of the detection element, it is apparent that the external stress is short in the cross section orthogonal to the axis O of the detection element. The rigidity when external stress is applied in the hand direction is lower than the rigidity when external stress is applied in the longitudinal direction. For this reason, the gas sensor using the sample having the chip on the short side is more acceptable for the chip than the gas sensor using the sample having the chip on the long side of the cross-sectional shape of the detection element. Can be increased. Therefore, as in this modification, the volume reference value and the half volume reference value of each chip are compared using the virtual image S and the virtual image T, and at that time, on the longitudinal side of the cross-sectional shape of the detection element The halved volume reference value may be set so that only the sample having a chip is a nonconforming product. In this way, a sample having a chip that does not affect the axial deviation between the metal shell and the detection element can be made a standard-compliant product, and the yield can be improved.

  Further, in the present embodiment, the top and bottom surfaces of the solid sealing material 260 are photographed by reversing the top and bottom of the solid sealing material 260 in the reversing step, but the reversing step is omitted. Using two CCD cameras, the bottom surface 266 side may be photographed while photographing the top surface 265 side of the solid sealing material 260, and the chipping that may occur on each surface side may be detected. In this case, a CCD camera is disposed on each of the upper side and the lower side (or the left side and the right side) of the inspection table as in the present embodiment, and the solid sealing material 260 is disposed on the top surface. Each of H.265 and bottom surface 266 is held on the inspection table in a state where it can be photographed by the CCD camera. For example, both sides may be photographed by holding the outer peripheral surface so as to float on the table, or the inspection table may be manufactured with a transparent plate and photographed with a CCD camera through the transparent plate. Then, if the top surface 265 and the bottom surface 266 of the solid sealing material 260 are photographed simultaneously or individually with two CCD cameras, and image analysis processing is performed in the same manner as in this embodiment, it is faster and more efficient than this embodiment, It is possible to detect chipping that may occur in the solid sealing material 260. In addition, when the above-described inspection table is used and photographing of the top surface 265 side of the solid sealing material 260 is finished, the position of the CCD camera 220 is moved to the opposite side across the inspection table, and photographing of the bottom surface 266 side is performed. You may go.

  A method for holding a member housed in a metal shell, such as an oxygen sensor, a NOx sensor, or an HC sensor, using a powder material, and a manufacturing method for supplying the powder material as a solid powder material Applicable.

1 is a longitudinal sectional view of a gas sensor 1. FIG. It is a figure which shows the structure of the computer 230 and its peripheral device. FIG. 3 is a diagram showing a manufacturing process of the gas sensor 1. FIG. 3 is a diagram showing a manufacturing process of the gas sensor 1. FIG. 3 is a diagram showing a manufacturing process of the gas sensor 1. It is a flowchart of the main routine of a standard conformity judgment program. It is a flowchart of the image analysis process of a standard conformity judgment program. It is a figure for demonstrating the 1st, 2nd imaging | photography process. It is a figure which shows the image formed virtually by the image data which binarized the picked-up image of the sample in the position A of a focus. It is a figure which shows the image formed virtually by the image data which binarized the picked-up image of the sample in the position B of a focus. It is a figure which shows the image formed virtually by the image data which binarized the picked-up image of the sample in the position C of a focus. It is a figure which shows the image formed virtually by the image data which binarized the picked-up image of the sample in the focus position D. It is a figure which shows the silhouette image of the sample formed virtually by the binarized image data. It is a figure which shows the image formed virtually by the image data which binarized the shape of the missing part of the sample in the focus position. It is a figure which shows the image formed virtually by the image data which binarized the shape of the missing part of the sample in the position C of a focus. It is a figure which shows the image formed virtually by the image data which binarized the shape of the missing part of the sample in the focus position. It is a figure which shows the virtual images S and T which divided the image of the sample formed virtually by the binarized image data into two. It is a figure which shows the virtual images S and T which divided the image of the sample formed virtually by the binarized image data into two.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Gas sensor 10 Detection element 11 Detection part 26 Sealing material 50 Main metal fitting 206 Solid sealing material 231 CPU
260 Solid sealing material 261 Insertion hole 265 Top surface 266 Bottom surface

Claims (2)

A detection element having a detection portion for detecting a specific component in the gas to be detected on the tip side of the detection tube and extending in the axial direction in a rod shape or a cylindrical shape,
A metal shell that has a cylindrical shape and surrounds the detection element;
A gas sensor manufacturing method comprising: a powder material disposed between an outer peripheral surface of the detection element and an inner peripheral surface of the metal shell,
In the process of manufacturing the gas sensor, the powder material is supplied as a cylindrical solid powder material in which powder is preliminarily pressed while having an insertion hole through which the detection element is inserted.
A first imaging step in which imaging of a first surface which is a surface on one opening end side of the insertion hole of the solid powder material is performed;
A second imaging step in which imaging of the second surface which is the surface on the other opening end side of the insertion hole of the solid powder material is performed;
Image analysis processing is performed on the photographed image of the first surface of the solid powder material photographed in the first photographing step, and the size of the chip generated on the first surface side of the solid powder material is detected. A first image analysis step,
Image analysis processing is performed on the captured image of the second surface of the solid powder material photographed in the second photographing step, and the size of a chip generated on the second surface side of the solid powder material is detected. A second image analysis step,
A first volume value calculating step in which a first volume value corresponding to the size of the chip on the first surface side of the solid powder material is calculated;
A second volume value calculating step in which a second volume value corresponding to the size of the chip on the second surface side of the solid powder material is calculated;
A standard conformity determination step in which the solid powder material is determined to be within a standard when both the first volume value and the second volume value are smaller than a predetermined volume reference value. A method for manufacturing a gas sensor. The detection element has a plate shape having a rectangular cross section perpendicular to the axial direction, and the opening shape of the insertion hole opened in the first surface and the second surface of the solid powder material is the detection element. It has a rectangular shape according to the shape of
In the first volume value calculating step, the first surface of the solid powder material is cut in the longitudinal direction of the opening shape of the insertion hole and divided in the lateral direction, and each half surface is divided by half. Two first volume values corresponding to the size of the chip included in the surface are calculated,
In the second volume value calculating step, the second surface of the solid powder material is cut in the longitudinal direction of the opening shape of the insertion hole and divided in the lateral direction, and each half surface is divided by half. Two second volume values corresponding to the size of the chip included in the surface are calculated,
In the standard conformity determination step, when all of the two first volume values and the two second volume values are smaller than a predetermined half volume reference value, it is determined that the solid powder material is within the standard. The method of manufacturing a gas sensor according to claim 1, wherein:

Images