JPH0776754B2 - IC lead bending detection method - Google Patents

Description

Detailed Description of the Invention

[Field of Industrial Application] The present invention relates to detection of lead bending of a flat package type IC for imposition, and is particularly suitable for use in combination with a visual sensor for automatically mounting such an IC on a substrate. IC lead bending detection method.

[Prior art]

Conventionally, for various electronic circuits, a structure using a printed wiring board has been widely adopted, but in recent years, except for the early days, such a board has been generally manufactured by an automated device. Therefore, various automatic IC mounting devices have been proposed and put into practical use. By the way, in such a case, if the lead of the IC to be mounted on the board is bent, the connection with the wiring pattern on the board cannot be accurately obtained. Bend should be inspected to ensure that only ICs in the correct lead state are brought to the board. Here, the bending of the leads of this IC will be described with reference to FIG. 6. First, FIG. 6 (a) is a plan view of a flat package type IC, in which 1 is the body of the IC and 2 is Represents a lead. And, 2a and 2b are leads having a bend. Further, FIG. 2B is also a side view, in which leads 2c and 2d are bent leads. In addition, hereinafter, the bending of the lead 2 shown in FIG. 6 (a), that is, the bending along the arrangement direction of the lead rows is called lateral bending, and the bending of the lead shown in FIG. 6 (b), that is, The bending of the normal rainfall, which is perpendicular to the arrangement direction of the lead rows, is called the vertical bending. On the other hand, in the automatic IC mounting device, an IC hand to be mounted on the board, which is prepared in advance at a predetermined location, is provided with a suction hand for transporting and positioning on the board.
After holding the IC with the suction hand, temporarily stop this suction hand at a predetermined position,
The holding position of the IC is detected, and the positioning of the suction hand on the substrate is controlled based on the detection result, so that the mounting accuracy of the IC can be sufficiently obtained. Then, in order to detect the holding position at this time, there is a method in which the IC held by the suction hand is imaged by a TV camera to obtain image data, and the holding position of the IC is detected from the image data. Known from the past. Therefore, various methods are conceivable in which this image data is used to detect the bending of the lead in the same manner. As an example of these methods, the image data described above is binarized, the center of gravity of each lead is obtained from this binarized data, the distance between the centers of gravity is detected as the lead pitch, and a preset value is set. It is considered that the detection is performed by comparing with. However, when the target IC has a narrow lead pitch and a large number of leads, in order to apply the method by binarization, a considerable resolution is required as the image data. Image data needs to be captured and image data with many pixels is required.
There is a problem in that it is necessary to perform multiple imaging operations for each IC, and the data processing time increases, resulting in a long board mounting processing time. It should be noted that related to this kind of technology, for example,
Reference can be made to JP-A-60-1900.

[Problems to be Solved by the Invention]

In the above-mentioned conventional technique, since the binarized image data processed as white or black or 1 and 0 is used, regarding the edge portion of the lead, the intermediate data which is neither white nor black is white instead of being the intermediate portion. The position of the detected edge is liable to be unstable because it is determined to be black and processed.
It is difficult to measure minute bends with a high lead pitch of about 0.3 mm width, which is generally used for IC leads, with high precision. Therefore, as described above, image data with a large number of pixels due to the expanded field of view is required. Therefore, the application of the IC to the automatic mounting device causes a problem that the processing time per IC increases. An object of the present invention is to sufficiently deal with the above-mentioned problems of the conventional example, and to use the image data for the automatic mounting device of the flat package IC for imposition without providing the image data capturing device exclusively for the bend detection. Another object of the present invention is to provide an IC lead bend detecting method capable of easily detecting lead bend at high speed and with high accuracy.

[Means for solving problems]

According to the present invention, according to the present invention, one-dimensional image data on a measurement line is set by setting a measurement line that crosses a row of leads in parallel with the arrangement direction of the leads from image data obtained by imaging the plane of an IC in one field The brightness data is extracted from, and the position coordinate of the lead edge is calculated from the extracted brightness data,
It is characterized in that the lead pitch is calculated from the coordinates of the adjacent leads and the lead pitch is detected by comparing a plurality of lead pitches. Further, by checking the number of pieces of data which are brighter than a predetermined value among the brightness data extracted from the one-dimensional image data, the number of leads can be grasped,
Whether or not the lead is bent can be detected by comparing the number of leads of the target IC.

[Action]

The lead line of the flat package type IC to be detected is
When it is imaged with reflected light, it has some contrast as a bright part with respect to the background, but the amount of reflected light changes depending on the angle of the reflective surface of the lead, and the contrast difference with the background becomes small. . Therefore, for lead rows with such characteristics, the brightness distribution on the measurement line (straight line) set so as to cross these lead rows along the array direction depends on the lead array state. And shows periodic changes. Therefore, the lead pitch can be easily detected by calculating the lead pitch from this change and comparing it with the adjacent lead pitch, or by checking the number exceeding the predetermined brightness from the change in brightness. I can.

【Example】

Hereinafter, regarding the method for detecting the IC lead bending according to the present invention,
This will be described in detail with reference to the illustrated embodiment. FIG. 2 shows an embodiment of the present invention, in which 1 is a flat package type IC, 3 is a suction hand, 4 is a chuck for rough positioning, 5 is a television camera, 6 is a lens, 7 is an image sensor, 8 is a lighting device. In this embodiment, the present invention is applied to an automatic IC mounting device. The suction hand 3 is provided with a chuck 4, and the IC 1 prepared in advance at a predetermined position is sucked and held. This also serves to carry and position the substrate at a predetermined position which is previously installed at a predetermined position. Here, the chuck 4 serves to roughly position the suction attitude of the IC 1 with respect to the suction hand 3. Then, at this time, the suction hand 3 is carrying the IC1 on the substrate from the above-mentioned predetermined position, as described above.
It temporarily stops at a predetermined position. On the other hand, the television camera 5 is installed right below the suction hand 3 corresponding to the position where the suction hand 3 is once stopped, and here, the lens 6 is an image in which the entire image of the plane of the IC 1 is included in the visual field. Is just imaged on the image sensor 7. Then, when the suction hand 3 has just stopped, the IC 1 in the suction holding state is imaged, the image data is fetched, and the image data is stored in a memory (frame memory) not shown. Then, by processing the image data stored in this memory, the suction position and the posture of the IC1 with respect to the suction hand 3 at this time are detected, and the positioning correction control on the substrate is performed thereafter, and at the same time, this embodiment is performed. Then, by further processing the image data stored in this memory, the bending of the lead 2 of the IC1 is detected, and it is determined whether or not the IC is a defective product. Control to remove. Next, detection of bending of the IC lead by this image data processing will be described. As already explained in Fig. 6, the bending of the IC leads is
Horizontal bending (Fig. 6 (a)) and vertical bending (same (b))
There is. Therefore, first, detection of lateral bending will be described. When the image of the IC imaged as shown in FIG. 2 is displayed on the screen of the CRT (not shown), it becomes as shown in FIG. Therefore, the measurement line L is set in advance at a predetermined position on this screen, and the brightness data D is extracted along this measurement line L. The brightness data D is one-dimensional image data, but if it is shown two-dimensionally for the sake of explanation, it has a pattern as shown in FIG. First, calculation of IC lead coordinates will be described. As for the calculation method, the invention of the same applicant, “Automatic mounting method for imposition components” (Japanese Patent Application No. 60-224852) and “Device for mounting IC on printed circuit board” (Japanese Patent Application No. 61-113586). Is also described. From the image data arrayed and stored in the frame memory in a grid pattern, the data D of the portion corresponding to this measurement line L is weighted averaged according to the following formula, and the value is taken as the edge coordinate Xk of the lead 2. And However, Xk: Edge coordinate value in the direction of measurement line L m: Brightness density monotonous increase, decrease start point n: Brightness density monotone increase, decrease end point f (i): Brightness data at i coordinate Monotonic increase at this time The decrease range is a range having the brightness change of the same direction component, and is determined by the condition of the following expression (2). | f (n) −f (m) | ≧ Dth (2) However, Dth: brightness threshold value At this time, the edge coordinates obtained by the above equation (1) are
It is detected with high accuracy in units of one pixel or less. The above is also shown in the above invention by the same applicant. The present invention will be described below. The lead pitches Lp and Lp + 1, which are the distances between adjacent leads, are found from the edge coordinates found by equation (1), and the presence of lead bends is detected by this, and the IC
The pass / fail judgment is made. The determination at this time is performed by the following (3), and when the condition of the expression (3) is not satisfied, it is determined as a defective product. | Lp−L (p + 1) | <Nth (3) However, Nth: value that is not judged L ₁ : 1/2 | X ₃ ＋ X ₄ −X ₁ −X ₂ ｜ ：: Lp: 1/2 | X (k + ₂ ) + X (k + ₃ ) -Xk-X (k + ₁ ) | L (p + ₁ ): 1/2 | X (k + ₅ ) + X (k + ₄ ) -X (k
_{+ 2) -X (k + 3} ) | Thus, according to this embodiment, as described above, the absolute value of the difference between the lead pitch between adjacent leads | Lp-L (p
+ ₁₎ | by determining, it is possible to determine the magnitude of the lead bending, the image data by the imaging at the time of automatic mounting process to the substrate of the IC used as it is, is possible to easily lead bending detection it can. Next, the detection of the vertical bending will be described. Returning to FIG. 2, in this embodiment, the illumination device 8 is arranged in the vicinity of the imaging lens 6 so as to surround the lens and is continuous. A ring light source device 8 using an optical fiber is used so that light is emitted upward from the light source array, and the IC 1 is imaged under illumination by the ring light source device 8. Then, the image formation state of the lead 2 by the optical system of this embodiment is shown in FIG. Here, 10 is a light irradiation port formed by an optical fiber ring of the illuminating device 8, 11 is an optical center of the lens 6, 12 is an optical axis of the lens, 13 is an image pickup surface of the image pickup element 8, 14 is a front principal point of the lens 8, 15 Is also a posterior point. First, the light rays α and β emitted from the light exit 10 are specularly reflected on the back surface of the lead 2, and the reflected light passes through the lens 6 and the imaging surface.
It reaches 13 and forms an image there. Therefore, the brightness of the image of the lead 2 on the image pickup surface at this time is the angle Θ of the lead 2 from the horizontal plane, that is, since the irradiation port 10 is sufficiently small and can be regarded as a point light source. It changes depending on the size of the bend of the lead 2. Then, at this time, the angle Θ for the lead image on the imaging surface 13 to have the maximum light amount is determined by the following equation. Θα = 1/2 [2π-Tan ^-1 d / a-Tan ^-1 b / (a + c)] ...
(4) Θβ = 1/2 [π-Tan ^-1 d / a + Tan ^-1 (ac) / b] ...
(5) where a: Distance from lens optical axis to lead reflection surface (point) b: Distance from irradiation opening to lead reflection surface (point) c: Distance from lens optical axis to irradiation opening d: Lead Distance from the reflecting surface (point) to the optical axis of the lens Therefore, according to this embodiment, appropriate values can be given as the constants a, b, c and d in the expressions (4) and (5). By determining the configuration of each part, the angle Θ
Is greater than the predetermined value, only the amount of light reflected by the IC that should be determined to be defective due to bending of the lead can be set to a predetermined value or less, and vertical bending can be detected. This detection is performed by the brightness threshold Dt in the above formula (2).
This is done by setting h to a predetermined value.
The number of times of the light amount portion showing the brightness of th or more is compared with the number of leads given in advance, and if they match, it can be determined as a good product. As is clear from the above description, the above-mentioned decrease in the amount of light also appears when the leads shown in FIG. 7 are not present. Therefore, according to this embodiment, it is possible to simultaneously obtain a pass / fail judgment for an IC in which an originally intended lead is lost due to a defect or the like. FIG. 5 shows a schematic flowchart of the lead bending detection processing in the above embodiment. First, in the process 500, the process of transferring the data along the measurement line L of a certain lead group from the image data stored in the frame memory to a predetermined buffer is performed. In processing 504 and processing 506, the edge coordinates are detected based on the above equations (1) and (2) according to the determination in processing 502, and the results are sequentially tabulated and the processing is repeated. In process 508, the inter-pitch distance calculation process is performed based on the above equation (3), and the result is determined in the next process 510, and lateral bending is detected. In process 512, a process for positioning control of the IC with respect to the substrate, which is originally required by the automatic mounting apparatus to which this embodiment is applied, is performed. In process 514, it is determined whether or not the above process has been completed for all read columns of the IC. That is, the IC generally has a plurality of lead rows, for example, four rows in the IC shown in FIG. Therefore, the processing is repeated for each measurement line corresponding to the plurality of lead rows. In general, most of the lead rows are 2 rows and 4 rows. In process 516, the change in the light amount obtained by the equations (4) and (5) is discriminated by the threshold value of the equation (2), and the number of leads is checked based on the discriminant, and the pass / fail judgment process is performed. In process 518, based on the result obtained by the above process 512, necessary for positioning control when mounting the board,
The holding position and posture of the IC suction hand 3 (Fig. 2) are detected. Therefore, according to this embodiment, when the IC automatic mounting process is performed, the lead bending can be detected only by sharing the image data used for the control in this process. As a result, lead bending can be reliably detected with almost no increase in the time required for a series of processes required to mount one IC on the substrate, that is, the so-called tact time. In the above embodiment, one measurement line L is set for each lead row of the IC as shown in FIG. 3, but instead of this, a plurality of measurement lines L is set for each lead row. Is set, the measurement is performed multiple times based on these multiple measurement lines, and the final detection is performed using the data showing the maximum number of leads among the detection results of the data of these measurement lines. However, according to this method, the reliability can be further improved. By the way, in some ICs, the arrangement pitch of the leads is not uniform, and some of them have different pitches, as shown in FIG. Therefore, the present invention can be easily applied to such an IC by configuring the IC so as to ignore the detection result in the portion where the pitch is different. For this part that should be ignored, the corresponding data should be set in advance.
If you decide, you can easily set it, and you can easily implement it without any problems. [Effects of the Invention] As described above, according to the present invention, high reliability can be easily achieved by using image data obtained by a single image pickup with the entire plane of the IC to be inspected as a field of view. Since it is possible to detect the lead bend while keeping it, it is possible to sufficiently deal with the problems of the conventional technology and to apply it to an automatic mounting device to always perform reliable detection at high speed without increasing the tact time. An IC lead bending detection method that can be performed can be provided.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is an explanatory view of a detection principle in an embodiment of an IC lead bending detection method according to the present invention, FIG. 2 is an explanatory view of an image pickup operation in the same embodiment, and FIG. FIG. 4 is an explanatory view of a measurement line in the invention, FIG. 4 is an explanatory view of an optical system at the time of image pickup in the same embodiment, FIG. 5 is a flowchart for explaining the operation in the same embodiment, and FIG. 6 is an example of an IC package. And FIG. 7 is an explanatory view showing another example of the IC package. 1 ... IC, 2 ... Lead, 3 ... Suction hand, 4 ... Chuck, 5 ... TV camera, 6 ... Lens, 7 ... Imaging element, 8 ... Illumination device,

Claims (2)

[Claims] 1. A method for detecting bending of a lead of an IC from image data obtained by picking up an entire plane of an IC to be inspected, wherein the lead in the image data is arranged along the arrangement direction of the lead. Setting a measurement line that crosses the column, extracting the brightness data from the one-dimensional image data obtained from the setting line, and calculating the position coordinates of each edge of the lead based on the change in the brightness data, The lead pitch, which is the distance between adjacent leads, is calculated from the position coordinates of the edge for each lead, and the lead pitch for each lead changes beyond a predetermined value set between adjacent leads. IC bending detection method characterized by determining that there is a bending in the lead. 2. The method according to claim 1, further comprising a process of calculating the number of changes of the brightness data from the brightness data and comparing the number of changes with a predetermined value set in advance. IC lead bending detection method.