JP2786859B2 - Electronic component mounting method in electronic component automatic mounting machine - Google Patents

Description

The present invention relates to a method for mounting an electronic component in an electronic component mounting machine, and particularly to an electronic component having an array of terminal pins such as an IC package (integrated circuit element). The present invention relates to a method of setting a line segment (measurement line) used to execute a measurement of a position or a posture of a contact point on a substrate using a visual sensor such as a camera by a change point detection method.

[Conventional technology]

The electronic component mounting machine is a device for automatically mounting a flat package IC as shown in FIG. 4, for example, on a substrate. At this time, the position of the IC package is controlled to match the contact row of the substrate. For this reason, measurement of the position of the contact point,
It is necessary to measure the holding position in a state where the IC package is held by the mounting suction hand.

Conventionally, there has been known a method of measuring a position of a contact point and a position of an IC package using a change point detection method using a visual sensor, but this requires setting of a measurement line.
This measurement line is a pin row protruding from the IC package,
Alternatively, a contact row on a substrate is displayed on an image as a pattern in which rectangular micro-rectangles are arranged in a straight line in a direction perpendicular to the longitudinal direction, and a pattern (image pattern) composed of a plurality of micro-rectangles arranged in this row Is set across.

At this time, in order to improve the measurement accuracy, practically, a plurality of measurement lines are set.

Incidentally, such a prior art is disclosed in Japanese Patent Application Laid-Open No. 63-1847.
There is the invention of Japanese Patent Publication No.

[Problems to be solved by the invention]

As shown in the prior art, when the center of gravity of a pattern displayed on an image of a pin array or a contact point is determined by a change point detection method, as shown in FIG. Draw a straight line that crosses the pattern consisting of, measure the luminance on this straight line, find the boundary position of the minute rectangle by detecting the change point as shown in FIG. 6, find the center of gravity of each pattern by averaging the load, etc. Seeking the center of gravity.

By the way, a flat package IC as shown in FIG. 4 often has pins in four directions. In this case,
Since pins are present on each side of the package body, at least four measurement lines are required.

Here, in order to calculate the position of the center of gravity, the accuracy increases as the number of measurement lines to be set in the pattern increases, but as shown in FIG. 9, the accuracy increases gradually as the number of measurement lines increases. When the number exceeds a certain number, the accuracy becomes substantially constant (hereinafter, this value is referred to as a saturation threshold), and the improvement in accuracy by increasing the number of measurement lines is very slight.

Since it is necessary to calculate the position of the center of gravity with high accuracy, it is desirable that the number of measurement lines is large, but if it is large, the measurement time increases.

On the other hand, from the viewpoint of shortening the measurement time, it is desirable that the number is small. However, if the number of measurement lines is smaller than the saturation threshold, there is a problem that the reliability is reduced. The number of the measurement lines becomes an appropriate value.

In the above prior art, the number of measurement lines is determined by an operator.

However, factors affecting data accuracy include:
The number of pins in the contact pattern, the distance between the pins, the brightness of the pattern, etc., determine the number of measurement lines by the operator. I was seeking.

An object of the present invention is to target an object such as an IC having pins at points along at least two sides of a main body having a substantially square planar shape, and arranged in a row when projected on a screen of a visual sensor. Automatic electronic component mounting machine that can set the number of measurement lines for determining the center of gravity of pins and contacts displayed as a pattern consisting of a plurality of minute rectangles to the minimum number of lines where the accuracy of the determined center of gravity is the desired accuracy. The present invention provides a method for mounting an electronic component.

[Means for solving the problem]

An object of the present invention is to mount an electronic component having a plurality of pins protruding from a side of a main body having a substantially square planar shape such that the pins are fitted to a plurality of contact patterns on a substrate. In the image obtained by imaging the electronic component with a visual sensor, the image pattern of each of the row of pins and the row of contact patterns is substantially orthogonal to the image pattern. A plurality of measurement lines including one above are set respectively, and the positions of the pin rows and the contact pattern rows and the positions of the centers of gravity thereof are obtained by changes in the brightness of the image on the measurement lines, and the positions of the centers of gravity are determined. In the method of mounting an electronic component in an electronic component automatic mounting machine in which the electronic component is mounted on a substrate so that The average value of the center-of-gravity position when measured using the measuring line of a book with a maximum value of number of rare and have pixels as U _a, using N (<a) present in the measurement line, from N = 1 N = until a, when the average value of the center-of-gravity position when measured sequentially increasing the number of measurement lines was U _N, (U _a -
When the absolute value of U _N ) becomes equal to or smaller than a preset value ε, N is stored as the optimal number of measurement lines, and thereafter, mounting of electronic components is performed using the optimal number of measurement lines. Is achieved by

[Action]

Set many at the minute interval set on the pattern,
The position of the center of gravity of the pattern is calculated using a measurement line orthogonal to all the patterns. By setting a large number of measurement lines at minute intervals, the position of the center of gravity of the pattern can be obtained with high accuracy.

Next, an appropriate number of measurement lines are set at minute intervals set on the pattern, the center of gravity of the pattern is calculated, and the center of gravity determined with high accuracy by a large number of measurement lines set at minute intervals earlier is calculated. Is calculated.

If the shift amount exceeds a desired value, the number of measurement lines is increased, and the minimum number of measurement lines at which the shift amount is equal to or less than the desired value is obtained. I do.

Thereafter, for the same pattern, the measurement is performed by setting the optimum number of measurement lines stored in the storage means.

〔Example〕

Hereinafter, embodiments of the present invention will be described in detail with reference to FIGS. 1 to 8.

FIG. 1 is an external view of an electronic component automatic mounting machine. In the figure, reference numeral 1 denotes an orthogonal robot, 2 denotes a mounting head provided on the orthogonal robot 1, and a hand for mounting IC components on a substrate and a substrate. It has a camera that recognizes contacts. Reference numeral 3 denotes an IC component supply device, and reference numeral 4 denotes a control device.

The basic operation will be described with reference to FIG. 2. First, the substrate is supplied by the substrate transfer device 5 and mechanically fixed.
Next, the camera 7 attached to the mounting head 2 is moved on the substrate, and the position of the center of gravity is obtained by recognizing the contact pattern on the substrate. Next, the hand 8 of the mounting head 2 grasps the component supplied from the IC component supply device 3 and the IC component detection camera 10.
Move up to detect the component displacement. The IC is mounted on the substrate so that the center of gravity of the contact pattern on the substrate matches the center of gravity of the IC.

FIG. 3 is a hardware block diagram focusing on the visual recognition control unit.

Reference numeral 11 denotes a CPU (Central Processing Unit) which performs processing related to the visual sensor and feedback processing. Next, 12 is ROM, CPU
A program indicating the processing procedure performed in step 11 is stored. Reference numeral 13 denotes a RAM, which stores the progress of the processing of the CPU 11 and the like. 14 is a DP-RAM (dual port RAM), which can be accessed from the CPU that controls the mounting operation.
Using the P-RAM 14, the visual sensor unit and the control unit of the mounting operation exchange information.

Reference numeral 19 denotes an imaging control unit which, when the CPU 11 instructs imaging, instructs the bus switch 15 to disconnect the image memory 16 from the internal bus 17 and connect it to the image bus 18. Next, a synchronization signal is supplied to the camera 7 or the camera 10 as a visual sensor, and an A / D conversion command signal synchronized with the synchronization signal is sent to the A / D converter 20. The image signal digitized by the A / D converter 20 is
It is stored in the image memory 16 at the address specified by the MA control unit 21. This is performed for one screen, and the image is stored in the image memory 16.

The IC to be mounted is, for example, a flat package IC.

FIG. 4 shows an example of an image obtained when the pin array of the IC is captured by the camera 10.

On the image, the pin array has a pattern in which a plurality of elongated small rectangles are arranged in a straight line in a direction perpendicular to the longitudinal direction. Reference numeral 22 denotes a measurement line. In this example, three lines are shown for each pin pattern in one row, as shown in the figure.

Here, a method of detecting the position of the pattern from the image of the camera will be described with reference to FIGS.

As shown in FIG. 4, a measurement line 22 is set in a direction orthogonal to the longitudinal direction of each minute rectangle as shown in FIG. If the position of the measurement line 22 is fixed on the screen, the position of the pattern slightly shifts for each substrate or IC, but it can be mechanically positioned to such an extent that the measurement line 22 always intersects each minute rectangle of the pattern.

At this time, the brightness of the image on the measurement line 22 becomes liquid corresponding to the pattern arrangement as shown in FIG. 5, and from this waveform, the boundary position of each minute rectangle of the pattern on the measurement line 22 is obtained. Is required. In the fifth diagram of an example, the boundary X _i with the background for each small rectangle is required.

Next, a method of finding the boundaries of each minute rectangle of the pattern will be described with reference to FIG.

First, when the lighting condition is good and the brightness of the screen is uniform (FIG. 6 (I)), the brightness threshold is appropriately determined, and the positions of the pixels at both ends of the brighter or darker range are determined. Is determined as the boundary position of each minute rectangle.

When the brightness of the screen is substantially equal to FIG. 6 (I) (FIG. 6 (II)), the brightness of the pixels on both sides sandwiching the threshold value is proportionally distributed from the brightness gradient of the pixels on both sides. If the position serving as the threshold is determined by a real value including a decimal point below the pixel position, the position can be determined with higher accuracy.

Further, when the brightness of the screen is not uniform due to the lighting conditions or the like (FIG. 6 (III)), the boundary position of each minute rectangle cannot be separated by one threshold value. Methods such as comparing the brightness and finding the position where the brightness gradient becomes the maximum as the boundary position, or calculating the weighted average A / B of the brightness f (x) of each pixel for each range from the peak to the valley What should I do?

 However, A and B are as follows.

As described above, the boundary position of each minute rectangle of the pattern in the measurement line direction is represented as a coordinate sequence by {X _i | i = 1,.
Then, the position of the center of gravity of the entire pattern in the measurement line direction is obtained by the following equation (3).

The above is described in Japanese Patent Application Laid-Open No. 63-18473, etc., and is called a change point detection method.

Generally, at least two measurement lines 22 are set in order to accurately determine the position of the center of gravity.

Since the position of the center of gravity in the direction of the measurement line on each measurement line is obtained by the equation (3), the average value of the positions of the centers of gravity is given by the following equation (4) when the number of the measurement lines 22 is N.

U _N = (U ₁ ′ + U ₂ ′ +... + U _N ′) / N... (4) When the center of gravity of each pattern in the measurement line direction is thus obtained, The position of the center of gravity of the pattern of the pin row or contact row is obtained.

The appropriate number of measurement lines is generally about 3 to 5, but if the number is small, the necessary mounting accuracy cannot be obtained.
If the number is too large, the measurement time will be long and the mounting time will be long, so that the minimum number of measurement lines is the optimum value within a range where the desired accuracy can be obtained. Therefore, when performing the electronic component mounting operation, the number of the measurement lines is determined in advance for each component or contact row pattern (hereinafter, an optimum value of the number of measurement lines is determined in advance for each component or contact row). Is referred to as teaching), and the optimum value of the measured number of measurement lines is stored for each pattern, and when mounting the IC on the board, the center of gravity of each pattern is determined by the number of measurement lines obtained at the time of teaching. Perform measurement.

Hereinafter, a method of obtaining the number of measurement lines at the time of teaching (step 81) will be described with reference to FIG.

First, the reference line 23 shown in FIG. 7 is set. The setting of the reference line 23 is performed by creating a rectangle with two points given on the screen as diagonal vertices, and using each side of the quadrilateral as a reference line of the measurement line.

For this reference line, the number of minute rectangles on the reference line is determined (step 82). This number of minute rectangles can be easily obtained from the number of position coordinate strings of each minute rectangle of the pattern described above.

Next, the measurement line 22 is set in line symmetry parallel to the reference line 23 at an appropriate minute interval.

The minute interval is an interval that is appropriately set, and may be set, for example, every one pixel or every two pixels based on the pixel.

Next, the number of minute rectangles on the measurement line 22 is calculated (step 83), and the number of minute rectangles on the reference line 23 is compared with the number of minute rectangles on the newly set measurement line 22 (step 8).
4) If the number of minute rectangles is equal, a new measurement line 22 is set symmetrically with respect to the reference line 23, and the number of minute rectangles on this measurement line 22 is obtained (step 85).

In this way, the measurement lines 22 are increased one by one at the same minute interval, and the measurement lines 22 are set until the number of minute rectangles on the reference line 23 and the newly set minute rectangle on the measurement line 22 are different.

When the measurement line 22 is set to a position outside any one of the minute rectangles of the pattern, the number of minute rectangles on this measurement line differs from the number of minute rectangles on the reference line 23.

Therefore, the number of measurement lines up to one before which the number of minute rectangles differs, that is, the number of measurement lines set so as to be orthogonal to all patterns at a set (minute) interval is set on the minute rectangle of the pattern. This is the maximum number of measurement lines 22 that can be obtained.

 Therefore, the maximum number of the measurement lines is set to a.

At this time, as described above, since the measurement line 22 is set for each pixel or every two pixels that constitute a minute rectangular image, the maximum number a eventually becomes very small for each pixel. The number is the same as the number of pixels included in the rectangular video, and in the case of every two pixels, it becomes half.

Then, then the average value U _a center of gravity position in the measurement line direction of the a present measurement line 22 (step 86) of a but, average value U _a of the position of the center of gravity, the above equation (4) Can be obtained by setting the number N in the above to the number a.

Then, the reference line 23 as a measuring line, measuring line 22 is determined center of gravity position U ₁ when the one (step 87), then compares these gravity center position U ₁ and U _a, which errors It is determined whether the value falls below the desired value (step 88). When the error of the gravity center position U ₁ and U _a exceeds a desired value, the measuring line 22 adds one, the average value U ₂ of the center-of-gravity position with respect to the two measurement line 22, and U _a Is compared.

In this way, the number of the measurement lines 22 is increased one by one (step 89), and the number of the measurement lines 22 that falls within a desired error is obtained.

Here, assuming that the desired saturation threshold is ε, this saturation threshold ε is 1 which is the minimum unit of the visual recognition device (visual sensor).
Assuming that the pixel accuracy is α and the mechanical accuracy of the electronic component mounting machine is β, ε = β / α.

Then, from the saturation threshold ε, N measurement lines satisfying the following expression (5) are obtained, and the N lines are the optimum value of the number of measurement lines.

| U _a −U _N | <ε... (5) Repeat steps 82 to 89 to determine the number of measurement lines 22 that fall within the desired error for each side (step
90). Further, the number of measurement lines 22 that can obtain the desired mounting accuracy with respect to the pins of all types of mounted ICs and the contacts on the substrate is obtained (step 91), and the teaching is terminated (step 92).

Based on the data on the number of measurement lines for each pattern obtained by teaching, the number of measurement lines is automatically set when components are mounted.

As described above, since the measurement line can be set and set to the lowest value for obtaining the desired accuracy for each pattern, the reliability of mounting and the mounting time can be shortened.

〔The invention's effect〕

According to the present invention, a package IC having pins at least along two sides of a main body having a substantially square planar shape
Target objects such as, when projected on the screen of the visual sensor, the number of measurement lines to determine the center of gravity of the pins and contacts displayed as a pattern consisting of small rectangles arranged in a row, Since the accuracy is set to the minimum number of the desired accuracy, the reliability of mounting and the shortening of the tact time can be obtained.

[Brief description of the drawings]

FIG. 1 is an external view of an electronic component mounting machine, FIG. 2 is an overall view of the electronic component mounting machine, FIG. 3 is a hardware block diagram of a visual recognition device control unit, and FIG. FIGS. 5 and 6 are graphs showing the relationship between the brightness of an image and the pattern position for explaining a method of obtaining coordinates corresponding to the pattern, and FIG. 7 is a pattern diagram showing a method of setting measurement lines. FIG. 8 is an enlarged view, FIG. 8 is a flowchart showing a method for setting the number of measurement lines, and FIG. 9 is a graph showing the relationship between the accuracy of the number of measurement lines. DESCRIPTION OF SYMBOLS 1 ... Orthogonal robot, 2 ... Mounting head, 3 ... IC component supply device, 4 ... Control device, 5 ... Board transfer device, 7
…… Camera, 8 …… Suction hand, 10 …… Camera, 11 ……
CPU, 12 ROM, 13 RAM, 14 DP-RAM, 15 Bus switcher, 16 Image memory, 17 Internal bus, 18
Image bus, 19 ... Imaging control unit, 20 ... A / D converter, 21 ...
... DMA control unit, 22 ... measurement line, 23 ... reference line, 24 ... pattern.

Continuation of the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G01B 11/00-11/30 G06F 15/62 380-420,15 / 66-15/68 H05K 13/04

Claims (1)

(57) [Claims] An electronic component having a plurality of pins protruding from a side of a main body having a substantially square planar shape is mounted so that the pins are fitted to a plurality of contact patterns on a substrate. Of the images obtained by imaging the electronic component with a visual sensor,
For each image pattern of each of the rows of pins and each of the rows of contact patterns, a plurality of measurement lines including one are set on the image so as to be substantially orthogonal to the image patterns. The position of each of the pin rows and each of the contact pattern rows and the position of the center of gravity thereof are determined by a change in the brightness of the image on the line, and the positions of the centers of gravity are determined. In the electronic component mounting method in the electronic component automatic mounting machine, the number of pixels included in the length direction of the image of the pin or the contact pattern is measured using a number of measurement lines having a maximum value. the average value of the center-of-gravity position when the U _a, N _(<a)
When using this measuring line, from N = 1 until N = a, a mean value of the center-of-gravity position when measured sequentially increasing the number of measurement lines was _{U N, (U a -U N} ) N is stored as the optimal number of the measurement lines when the absolute value of is equal to or less than a preset value ε, and thereafter, mounting of the electronic component is performed using the optimal number of the measurement lines. Of electronic components in electronic component automatic placement machines.