JPH03265199A - Alignment device - Google Patents

Abstract

PURPOSE:To load correctly a substrate with electronic parts irrespective of position deviations existing in both sides of the substrate and the electronic parts by providing a substrate position recognition part and a component parts position recognition part and the like and detecting each amount of respective position deviations in the substrate and the electronic parts. CONSTITUTION:A substrate position recognition part 5 which detects the position deviations Xb and Yb of a substrate 11 by images of an optical system 3 for photographing the substrate and a component parts position recognition part 6 which detects the position deviations Xp, Yp, thetap of the electronic parts 2 by a component parts position detecting means 6a and a slanting angle detecting means 6b are provided. Then correction value is calculated by each amount of detection Xb, Yb and Xp, Yp, thetap as well as each amount of drive errors of a component parts loading head 8 and an X-Y table 7 and then, the drive of the head 8 and the table 7 is corrected. In this way, the accuracy of adjustment is improved by comparing with conventional adjustment that is performed for either one side position deviation in the substrate 11 or the electronic parts 2 and then, more correct movement can be performed. Further, deviations in any slanting direction of the electronic parts are detected by the same control program and thus the capacity of this device can be minimized.

Description

[Detailed Description of the Invention] [Summary] The present invention relates to a positioning device that positions a multi-terminal (bin) electronic component at a predetermined location on a board, and is capable of accurately positioning the electronic component on the board regardless of misalignment on both the board side and the electronic component side. The purpose of the present invention is to realize an alignment device that can be mounted on a board, and has an optical system for photographing a board that photographs the pattern of the board, and detects the amount of positional deviation of the board based on the image from the optical system for photographing the board. a component position detection means having a board position recognition unit, a component photographing optical system for photographing a terminal portion of the electronic component, and detecting a positional shift amount of the terminal portion using an image from the component photographing optical system; a component position recognition unit that detects an amount of positional deviation of the electronic component using an inclination angle detection means that detects an inclination angle θ of the terminal portion without depending on the inclination direction using an image from a photographing optical system; The amount of positional deviation detected by the position recognition section, the amount of positional deviation detected by the component position recognition section, the component mounting head and the X-
an offset memory that calculates and stores a correction value based on the drive error amount of the Y table, and sends the correction value to the control unit so that the drive of the component mounting head and the X-Y table is corrected according to the correction value; It consists of

[Industrial application field]

The present invention relates to a positioning device for electronic components, and more particularly to a positioning device for positioning a multi-terminal (bin) electronic component at a predetermined location on a printed circuit board.

■With the rapid progress in the high integration of electronic components such as C, the miniaturization of the packages that form these electronic components and the increase in the number of bins have become important factors for high-density packaging.

Therefore, the conventional DIP type (Dual I n-1i
As a new package to replace the neP package), the PGA type (P package) shown in the perspective view of the package in Figure 7 is available.
InGrid Array) is now being applied to mounted components.

In such an electronic component 2, not only is the pitch L of the bins 2B where signals are input and output narrow and the number of bins 2B is large, but also the arrangement of the bins 2B cannot be visually observed from directly above the component. It is no longer difficult to mount the electronic component 2 on the board 11 manually and visually.

Therefore, when electronic components 2 are mounted on a substrate 11 such as a printed circuit board, the electronic components 2 are now automatically mounted using a component mounting machine that uses image recognition technology.

As shown in the perspective view of FIG. 8, such a component loading machine includes a container 1 in which electronic components 2 are set in a housing 20;
It is formed by disposing an XY table 7 on which a board 11 is loaded, and by providing a rad 8 for mounting components driven by a drive mechanism 8A.

Then, the electronic component 2 set in the container 1 is picked up by the component mounting head 8, and the component mounting head 8 is driven by the drive mechanism 8A, and the electronic component 2 is placed in the bin at a predetermined location of the pattern IIA provided on the board 11. 2B is positioned and electronic components 2 are mounted one after another.

Therefore, a positioning device is provided for such positioning, and when mounting the electronic components 2, a large number of bins 2B are used.
High precision is required to ensure accurate positioning.

[Conventional technology]

The conventional alignment device was constructed as shown in the conventional explanatory diagram of FIG. Figure 6 (al).

(bl) is a block diagram showing the configuration, (a2) is a configuration diagram of an optical system for photographing a board, and (b2) is a configuration diagram of an optical system for photographing electronic components.

As shown in FIG. 6(al), the head drive section 22 is controlled by the control section 21 to drive the component mounting head 8, pick up the electronic components 2 from the container 1, and load them onto the X-Y table 7. Electronic components 2 are placed at predetermined locations on the board 11.
It is configured such that the electronic component 2 is positioned and the electronic component 2 is mounted.

In this case, the component mounting head 8 is usually a suction head that suctions the electronic component 2 with a nozzle at its tip by suctioning air.

Further, the table driving section 2 controlled by the control section 21
3 is connected to a board position detecting section 24 that compares the image of the board 11 photographed by the board photographing optical system 3 with reference image data D1 prepared in advance. The XY table 7 is configured to detect the amount of positional deviation and adjust the position of the XY table 7 based on the detected amount of positional deviation.

Furthermore, the optical system 3 for photographing the board in this case is shown in FIG. 6 (a2).
As shown in FIG.
, passing through the light shielding mask 3C2 and the projection lens 3D
- Irradiate the substrate 11 loaded on the Y table 7 and pass the reflected light through the imaging lens 3E and shielding mask 3F for CC
It is configured to receive light by D3G and photograph a silhouette image of the pattern 11A provided on the substrate 11.

〈Regarding such an optical system for photographing substrates, please refer to the patent application filed in 1983.
There is No.-13440. ) Then, while the electronic component 2 is transferred from the container 1 to the board 11 by driving the component mounting head 8, position adjustment is performed by moving the board 11 in the X and Y directions, and the electronic component 2 is placed at a predetermined location. It has been designed to be installed in the

Moreover, such fIt adjustment can also be performed with the configuration shown in FIG. 6(bl).

That is, when the electronic component 2 is picked up by the component mounting head 8, it is photographed by the component photographing optical system 4, and this photographed image is combined with the reference image data D2 prepared in advance.
are compared in the component position detection section 28, and the amount of positional deviation is detected and notified to the head drive section 27, thereby adjusting the position of the component mounting head 8.

When the electronic component 2 is mounted on the board 11 loaded on the X-Y table 7, when the electronic component 2 is picked up from the container 1 by the component mounting head 8, the electronic component 2 is The component mounting head 8 is adjusted according to the amount of positional shift, and after this adjustment, the component mounting head 8 is driven to place the electronic component 2 in a predetermined position on the board 11. It is designed to be implemented in certain locations.

As shown in FIG. 6(b2), the component photographing optical system 4 moves the pin 2B of the terminal portion 2A of the electronic component 2 attracted by the nozzle at the tip of the component mounting head 8 to the laser generating element 4A. Irradiates from all directions with laser light,
The camera 4F is configured to take pictures.

In this case, the laser beam irradiation by the laser generating element 4A is performed on the side surface of the bottle 2B by the cylindrical lens 4B, the mirror 4C1, and the cylindrical lens 4D, and the irradiated bottle 2B is formed so as to be photographed by the camera 4F through the filter 4E. has been done. (For such an optical system for photographing parts, see Japanese Patent Application Laid-Open No. 63-27590.
There is Publication No. 7. > Therefore, when the electronic component 2 is picked up, the amount of positional deviation is checked based on the photographed image of the component photographing optical system 4, and if positional deviation is detected, the component mounting head 8 is
After adjusting the position, the electronic component 2 is transferred to the board 11 and mounted.

[Problem to be solved by the invention]

However, the configurations shown in FIGS. 6(al) and (bl) detect the positional deviation of either the board 11 or the electronic component 2. For example, the positional deviation of the board 11 is detected and the position is adjusted. Even if the electronic component 2 is actually misaligned, it will not be possible to mount it in the specified position, and conversely, even if the misalignment of the electronic component 2 is adjusted,
If there is a positional shift on the board 11, it cannot be mounted in a predetermined position.

Furthermore, if there is a device-specific drive error in the drive of the X-Y table 7 and the component mounting head 8 due to misalignment in the installation of the board photographing optical system 3 and the component photographing optical system 4, a predetermined position on the board 11 Even if the electronic component 2 is operated to be transferred at the same time, there is a problem in that the electronic component 2 may not be mounted at a predetermined location due to a driving error.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to realize a positioning device that allows electronic components to be accurately mounted on a board regardless of positional deviations on both the board side and the electronic component side.

[Means to solve the problem]

FIG. 1 is an explanatory diagram of the principle of the present invention for solving the above-mentioned problems, and includes a substrate photographing optical system 3 for photographing a pattern on a substrate 11, and an image from the substrate photographing optical system 3. It has a board position recognition unit 5 that detects the amount of positional deviation (Xb, Yb) of the board 11, and a component photographing optical system 4 that photographs the terminal portion 2A of the electronic component 2, and an image from the component photographing optical system 4. The amount of positional deviation between the two terminals (Xb, Yb)
electronic component 2 by component position detection means 6a that detects the electronic component 2, and tilt angle detection means 6b that detects the tilt angle θp of the terminal portion 2A without depending on the tilt direction based on the image from the component photographing optical system 4. positional deviation amount (X.

A component position recognition unit 6 that detects Y□θp) and a positional deviation amount (Xb, Yb) detected by the board position recognition unit 5
, the positional deviation amount (X□Y, θ,) detected by the component position recognition unit 6, and the drive error amount (dx, d□d,) of the component mounting head 8 and the X-Y table 7. The correction value (XY, θ) is calculated and stored, and the correction value (
The correction value (XY, θ) is adjusted so that the driving of the component mounting head 8 and the X-Y table 7 is corrected by the correction value (XY, θ).
) to the control section 9.

[For production]

In FIG. 1, a substrate 1 loaded on an X-Y table 7
1 is provided with a board photographing optical system 3 and a board position recognition section 5, and the electronic component 2 held by the component mounting head 8 is provided with a component photographing optical system 4 and a component position recognition section 6, respectively.
The amount of positional deviation (Xb, Yb) detected by the board position recognition unit 5 and the component position recognition unit 6 and (Xb, Yb,
θ, ) are sent to the offset memory 10.

However, in the present invention, the inclination deviation amount θp in the component position recognition unit 6 is detected by the inclination angle detection means 6b which can detect the inclination direction regardless of the inclination direction (plus direction or minus direction).

In the offset memory 10, the amount of positional deviation (X.

y, ), (Xp, Yp, θp), and the drive error amount (dx, d□d,) of the x-y table 7 and component mounting head 8 to calculate the correction value (X, Y, θn, The corrected values (X, Y, θ) are notified to the control unit 9, and the positional deviation of both the X-Y table 7 and the component mounting head 8 is adjusted.

Therefore, compared to the conventional positional deviation adjustment of either the board 11 or the electronic component 2, the accuracy of adjustment is improved, more accurate transfer can be performed, and quality can be improved through reliable mounting. In addition, since the tilt deviation amount θp can be detected regardless of the tilt direction of the electronic component 2, there is no need for separate detection calculation programs for the plus direction and the minus direction, and only one control program is required, reducing its capacity. It has become a thing.

〔Example〕

Hereinafter, the alignment device according to the present invention will be described in detail with reference to FIGS. 2 to 5 showing embodiments thereof.

FIG. 2 is a block diagram of an example of the present invention, FIG. 3 is a flowchart showing the operation of the present invention, and FIG. is a camera arrangement diagram, (b) is a density graph, (c) and (d) are illustrations of pixel count aggregation, and (e) and (f) are illustrations of bin arrangement. (g>, (j>) is a plan view of the window, (h) is an explanatory diagram of counting the number of pixels in the window, (k
) is a flowchart diagram showing the bin position shift recognition process,
FIG. 5 is an explanatory diagram of recognition of positional deviation of a substrate according to the present invention, in which (al) and (a2) are plan views of main parts of the substrate, (b) is a density graph, and (c) is a pixel. Explanation diagram for aggregating numbers, the same figure (
d) is a flow chart diagram showing a process for recognizing a positional deviation of a substrate. The same reference numerals indicate the same objects throughout the figures.

As shown in FIG. 2, the bin 2B of electronic components 2 picked up and held by the component mounting head 8 driven by the head drive unit 13 is photographed by the component photographing optical system 4, and the photographed object is photographed by the component photographing optical system 4. The image data DP is sent to a component position recognition unit 6 which includes a ROM 6B and a RAM 6C and is controlled by a CPU 6A.

Further, the substrate 11 loaded on the X-Y table 7 driven by the table driving section 12 is moved by the optical system 3 for photographing the substrate.
This photographed image data DP is C
ROM5B and RAM5 controlled by PU5A
It is sent to the board position recognition unit 5 which has a built-in PC.

The positional deviation data (x,,y,,θ,〉) calculated by the component position recognition unit 6 and the positional deviation data (XY, Yゎ) calculated by the board position recognition unit 5 are stored in the recording area of the offset memory 10. They are stored in E2 and E3 respectively.

Also, in the recording area E4 of the offset memory 10, error data obtained by measuring the mounting error of the board photographing optical system 3 and the component photographing optical system 4, and the driving error of the component mounting head 8 and the X-Y table 7 is stored. D3 (dx, d□d,) is stored, and these positional deviation data (Xp, Yp, θp) and (Xb, Yb) and drive error data (dx, d, ,
d, ), the correction values (x, y
, θ> are calculated, and the correction values (X, Y, θ) are sent to the control unit 9, and the head drive unit 13 and table drive unit 12
By controlling the component mounting head 8 and the XY table 7
The position is adjusted.

Further, the image data DP is input to the camera interface 6D, A/D converted, and stored in the frame memory section 6E. In this case, the zoom mechanism 4J of the component photographing optical system 4 is driven by the zoom controller 6F to increase or decrease the magnification of the photographed image, and the movable table 4H is moved by the movable table controller 6G to obtain an enlarged image of a predetermined location. It is now possible to

Then, positional deviation data (X, ,
Y□θp) is calculated by the hard calculation processor 6H, and the calculation result is sent to the offset memory 10 via the external interface 6J.

Similarly, image data DB is input to the camera interface 5D, A/D converted, and stored in the frame memory section 5E. Then, positional deviation data (X, Y, Y) is calculated by the hardware calculation processor 5F based on the image data DB, and the calculation result is sent to the offset memory 10 via the external interface 5G.

When a predetermined electronic component 2 is to be mounted on the board 11, as shown in FIG. Parts are set to be moved to a predetermined location in system 4 (step S1).

On the other hand, the X-Y table 7 is moved in accordance with the mounting data stored in the floppy disk 9A so that the position on the board 11 on which the electronic component 2 is mounted faces the board photographing optical system 3 (step S2).

Next, the position of the electronic component 2 is recognized by the component photographing optical system 4 and the component position recognition unit 6 (step S3),
The position of the board 11 is recognized by the board photographing optical system 3 and the board position recognition unit 5 (Stellar 7S4).

As a result, positional deviations due to recognition of the positions of the electronic component 2 and the board 11 are sent to the offset memory 10, and the offset memory 10 receives a correction value (X).

Y, θ) are calculated and stored (step S5).

Then, the component mounting head 8 and the X-Y table 7 are driven based on the correction values (X, Y, θ), position adjustment is performed, and component mounting is performed (step S6).

When the mounting of one electronic component 2 is completed, the presence or absence of the next electronic component 2 is checked by referring to the above-mentioned mounting data, and if "Yes", the component mounting is executed by repeating the above-mentioned operation again. "If there are no parts, the parts mounting process will end.

Therefore, when mounting the electronic component 2 at a predetermined location on the board 11, the positional deviation of the electronic component 2 is corrected by the correction values (X, Y, θ), and mounting is performed with accurate positioning.

Next, explanatory diagrams of FIGS. 4(a) to (j) and FIG. 4(k)
The component position recognition process (the process of step S3 in FIG. 3) will be explained with reference to the flowchart.

First, an initial imaging range is selected (step 511).
). That is, the zoom mechanism 4J is driven by the zoom controller 6F to reduce the magnification and widen the imaging range (field of view) so that the entire image of the electronic component 2 can be input.

The camera 4F of the optical system 4 for photographing parts is shown in Fig. 4 (a).
), the terminal section of the electronic component 2 to be photographed is placed at a location A and a location B so as to be perpendicular to the half mirror 4G, which is disposed at a position corresponding to the two people. The camera 4F provided at location B may have a wide field of view and low magnification, and the camera 4F provided at location B may have a narrow field of view and high magnification, and may be used by switching between A and B. In this case, it is possible to freely obtain an enlarged image of the entire terminal portion 2A or a portion thereof without providing the zoom mechanism 45 as described above.

Image data DP photographed by the parts photographing optical system 4
Recognition of the component position using the following procedure is performed in the component position recognition section 6 as the component position detection means 6a and the inclination angle detection means 6b.

First, image data DP from the component photographing optical system 4 is inputted through the camera interface 6D, A/D converted, and stored in the frame memory section 6E. And each concentration O~255
The number of pixels corresponding to is counted, and a density histogram as shown in FIG. 4(b) is created (step 512).

Next, a binarization slice level t is set (step 513). That is, since the total number of pixels in the entire component bin of electronic component 2 is known from the design value, the area closest to that value (the shaded parts in the figure) is calculated by integrating from the bright side of the density histogram, and then Let the density of t be the binarization slice level t.

Then, using this binarization slice level t, the image data stored in the frame memory section 6E is binarized (step 514). That is, the binarization slice level t
Binarize (A/D conversion) the darker image, that is, the background image, at 0゛, and the image that is brighter than the binarization slice level t, that is, the component bin image, at 1°, and convert this binarized image. Store the data in frame memory 6E.

Next, based on the data stored in the frame memory 6E, a vertical projection distribution as shown in No. 41N(c) is determined (step 515). It is obtained by adding each column on the axis.At this time, the X-axis coordinates (Xi, X2) at both ends of the distribution
And the length m (=X2-XI) of the distribution is also determined.

Similarly, based on the data stored in the frame memory 6E, a horizontal projection distribution as shown in FIG. 4(d) is determined (step S16). It is determined by adding each row on the axis. At this time, the Y-axis coordinates (Yl, Y2) at both ends of the distribution are also determined.

Next, the reference bin (bin 1 PI) is searched (step 517), that is, the intersection point (Xl.

Y2) is set as the origin P, and as shown in Fig. 4(e) and (f), scan the binary image from the origin P in the X-axis direction, and first search for the coordinates that correspond to the pixel at ゛ 1°. By doing so, the value of bl in the same figure is obtained.

Next, from the value of m obtained earlier, a + (a + = m
b) Calculate I) and find the inclination direction of the part using the following comparison formula.

When a + < b +: θ = 10 θ (taken as plus) a
When +>b+: θ=-θ (assumed to be negative) Since the expected maximum inclination deviation is ±10°, in this example, r a = b + J, that is, θ = 45°. is not assumed.

The position of the No. 1 pin P1 is recognized from the obtained inclination direction, al and su.

Next, the tilt angle is detected (step 518).

The above inclination angle θ can be found by +θ−jan (at/bz) −θ= tan (b +/a + ), but in this case, two calculation programs corresponding to the polarity of the inclination angle are prepared. Must.

Therefore, in the present invention, by using the following rotation angle calculation method, one calculation program is required regardless of the polarity of the tilt angle.

That is, the inclination angle is calculated by substituting the illustrated bin end design value e and the previously determined value of m into the equation for calculating θ derived by the following procedure.

■If θ is positive: From sin θ = a, yl, cos θ = b, yl,
sinθ+eO8θ=a,/1 +b,/1= (a,
+b, ) / 1 = m / 12 Also, from the addition theorem, sin θ + cos θ = I (θ + α〉t
Since anα=1, substitute α−45° to obtain path sin (θ+45′)=m, #. From this, sin (θ+45°) −m/ <1”I−1)θ
+45° = sin-' (m/ (-'τ・1))
Therefore, θ is as follows.

θ=sin-'(m/(,'?'l))-
45°・-(1)■When θ is negative: eosθ=a,/1, sinθ=b+/1, but the following is the same as when θ is positive, so θ is as follows. It becomes like this.

θ-5in-' (m/ (Iτ・1)-45°・,-
(2) Therefore, the calculation formula for θ is the same whether it is positive or negative, but after calculating the value of θ, a polarity sign is added according to the previously determined slope direction, and the calculation program is 1 That's all it takes.

Next, the window is set (step 5t9)6, that is, based on the pitch L which is the design value, taking into account the coordinates of the first pin P1 and the inclination angle θ obtained by the above processing.
A window Wl centered on the bin 2B separated by the dotted line shown in FIG. 4(g).

W2. W3. Set...

Next, an enlarged image is input (step S20>. That is, the field of view is narrowed by increasing the magnification of the camera 4F, and enlarged image data DP of each window Wl, W2, W3, etc. is input, /D conversion and stored in the frame memory section 6E.

Next, the enlarged image data is binarized (step 52).
1) This binarization is performed using the same method as that performed in step S14 above.

Next, the centroid point of each bin is detected (step 522).
That is, as shown in FIG. 4(h), the number of pixels in the X-axis and Y-axis directions in the window W of the enlarged image stored in the frame memory section 6E is totaled, and the vertical projection distribution and horizontal projection distribution are calculated. The center of gravity WG of the window W is calculated by calculating the center of gravity XG and YG where the hatched areas S1 and S2 of the vertical projection distribution and the horizontal projection distribution are divided into two.

Then, it is checked whether all bins 2B have been completed (step S23), and if not, the above operation is repeatedly executed until all bins have been completed (step S22.823>
.

Next, the posture of the parts is recognized (step S24). That is, as shown in FIG. and y2 are calculated by the least squares method, and yl and y2
Find the line &iy that divides the intersection point into thirds, and calculate the angle α between the straight line y and the X axis.

Then, coordinates (Xo, yo>) which are 1/2 of the diagonal line in the design value of the electronic component 2 are calculated on the straight line y.

Finally, the amount of component positional deviation is calculated (step 525).
), that is, the positional deviation data X, , Y, θ of the electronic component 2
, can be obtained by the following equations (3), (4), and (5).

X, = X, -X, -= (3) Y-= YOy ... (4) θ9 = 45-α ... (5) However, in this case, These are the coordinates when the deviation is zero, and are calculated values in advance.

Next, explanatory diagrams of FIGS. 5(a) to (c) and FIG. 5(d)
The component position recognition process (the process of step S4 in FIG. 3) will be explained with reference to the flowchart.

First, the image data DB photographed by the substrate photographing optical system 3 is input (step 331).

That is, as shown in FIG. 5(al), a pattern 11A corresponding to the pin 2B and a reference mark IIB are arranged at the location where the electronic component 2 is mounted on the board 11, and the pattern 11A and the reference mark IIB are arranged as an image data DB. The image of the reference mark 11B is shown in the same figure (a
2) Enter the following.

Next, a density histogram is created (step 53).
2) That is, in this image data DB, the number of pixels corresponding to predetermined densities 0 to 255 is counted to create a density histogram, as shown in FIG.

Next, a binarization level t' is set (step 533). In this case, the area of the reference mark 11B is known from the design value, and is found by integrating the number of pixels that are closest to that value from the brightest side of the histogram, and the density at that time is taken as the binarization slice level t'.

Next, binarization is performed using the binarization slice level t' (step 534), that is, the binarization slice level t
A/D conversion is performed by binarizing the image data DB by setting values darker than '0' to '1' and values lighter than binarization slice level t' to '1', and storing the result in the frame memory section 5E. .

Next, the center of gravity of the reference mark is detected (step 535).
). This converts the pixel 'l' of the image data DB to the same figure (c
) as shown on the X-axis and Y-axis to create a vertical projection distribution and a horizontal projection distribution. By calculating XG' and YG', the reference mark I
Calculate the center of gravity WG' of IB.

Next, the amount of substrate positional deviation is calculated (step 536
). The positional deviation data (xbY,) of the substrate 11 are the coordinates (Xu', Y,
') can be obtained by the following formulas <6) and (7).

X, =XG'-Xu' ...(6) Y, =YG'
-Y'...(7) Next, offset memory 1
o will be explained.

Because there is an error in the coordinate system due to mounting accuracy between the coordinate axes of the component photographing optical system 4 and the board photographing optical system 3 that detect the component position and the board position, respectively, and the coordinate axes of the mechanical system at the component mounting position. , these are usually corrected with respect to a reference (usually using the machine coordinate system at the component mounting position).

However, this has the following problems.

Problem ■: It is difficult and time consuming to measure the difference between each coordinate axis with respect to the standard.

Problem ■: Add two pieces of data for the difference in coordinate axes (on the component side and on the board side)
Not only does it require memory, but it also takes time to perform calculations.

Problem ■; If multiple similar devices are manufactured, the above problem ■
■ becomes more noticeable.

To solve such problems, offset memory 10
is provided.

This is done by preparing reference parts and boards created according to design values, and calculating the part position detection results (Xp).

The component is actually mounted using a composite value obtained by simply adding Yp, θp> and the board position detection result (X, Y, ). Here, the component position detection result is the amount of positional deviation of the component based on the coordinate axis of the optical system 4 for photographing the component, and this reference is an orthogonal coordinate passing through the center of the camera field of the optical system 4 for photographing the component. be. In addition, the board position detection results are obtained from the board photographing optical system 3.
This is the amount of positional deviation of the substrate based on the coordinate axis of , and this reference is an orthogonal coordinate passing through the center of the camera field of view of the optical system 4 for photographing the substrate.

Then, by actually measuring the amount of positional deviation between the pattern 11A of the substrate 11 and the electronic component 2, the following equations (8) and (9) are obtained.

Values with opposite signs such that each of x, y, and θ in (10) are zero are written in advance in the offset memory 1o.

X=Xp+X-10dx=O...'(8) Y=Yp+Y,
+dr=O−<9> θ=θp 10d e ” O...(10) Here, (X, Y, θ) are the correction values of the machine control system, and (X,
, Y1θP) is the value of the position recognition result when using the reference component, 〈X1Y,) is the value of the position recognition result when using the reference board, and (d x , d
y, da) are error values stored in the offset memory 10.

Note that on the board side, by recognizing the reference mark 11B provided at the corner of the board 11 in advance, there is no problem in terms of accuracy even if angle recognition is not performed for each mounting position.

[Effects of the Invention] As explained above, according to the present invention, when mounting an electronic component on a predetermined location on a board, positional deviation data (Xp, Yp, θp) of the electronic component and positional deviation data of the board are prepared in advance. (X,, Yb) and installation and drive error data <dyod,
, d,) are stored in the offset memory, and adjustment values (X, Y, θ) for correcting these positional deviations can be easily retrieved. Drive the Y table to eliminate positional deviation,
Not only can reliable mounting be performed, but also a single control program is required, especially since the tilt shift amount θp of the electronic component in the positional shift data can be detected regardless of the tilt direction.

Therefore, it is possible to improve the positioning accuracy of mounting, and to provide a positioning device that can detect the amount of deviation of the electronic component in any direction of inclination using the same control program and reduce its capacity.

[Brief explanation of drawings]

FIG. 1 is a diagram illustrating the principle of a positioning device according to the present invention, FIG. 2 is a block diagram of an embodiment according to the present invention, FIG. 3 is a flowchart diagram for explaining the overall operation of the present invention, and FIG. is an explanatory diagram of recognition of bottle positional deviation according to the present invention;
) is a camera layout diagram, (b) is a density graph diagram, (c),
(d) is an explanatory diagram of the total number of pixels, (e) and (f) are explanatory diagrams of the bin arrangement, (g>, (j> are plan views of windows, (
h) is an explanatory diagram of counting the number of pixels in the window, (k> is a flowchart diagram, and FIG.
l), (a2) are plan views of the main parts of the board, (b) is a density graph diagram, (c) is an explanatory diagram of counting the number of pixels, (d) is a flowchart diagram, and Figure 6 is a conventional explanatory diagram. (al>, (bl) are tank bottom block diagrams, (a2) is a photographed configuration diagram of the board, (b2) is a photographed configuration diagram of electronic components, Figure 7 is a perspective view of the package, Figure 8 is the component mounting machine In the figure, l is a container, 2 is an electronic component, 3 is an optical system for photographing a board, 4 is an optical system for photographing a component, 5 is a board position recognition unit, 6 is a component position recognition unit, and 6a is a Component position detection means, 6b is an inclination angle detection means, 7 is an X-Y table, 8
1 is a component mounting head, 9 is a control section, 10 is an offset memory, 11 is a substrate, 12 is a table drive section, and 13 is a head drive section.

Claims (1)

[Claims] An electronic component (2) set in a container (1), a head drive unit (13) that drives a component mounting head (8) in which the electronic component (2) is held, and X-Y table (7) loaded with the board (11) on which the component (2) is mounted
a table drive section (12) that drives the component mounting head (8), a control section (9) that controls the drive of the table drive section (12) and the head drive section (13); - The electronic component (2) is driven by the Y table (7).
) at a predetermined location on the substrate (11), the device includes a substrate photographing optical system (3) for photographing the pattern of the substrate (11), and the substrate photographing optical system (3) ) The amount of positional deviation (X_b, Y_b) of the substrate (11) is determined by the image from
); and a component photographing optical system (4) for photographing the terminal section (2A) of the electronic component (2), and an image from the component photographing optical system (4). The amount of positional deviation (X_p, Y
The terminal portion (2A) is detected by the component position detection means (6a) for detecting the component position (6a) and the component photographing optical system (4).
and a component position recognition unit (6) that detects the amount of positional deviation (X_p, Y_p, θ_p) of the electronic component (2) by the tilt angle detection means (6b) that detects the tilt angle θ_p of the electronic component (2) without depending on the tilt direction. ), the positional deviation amount (X_b, Y_b) detected by the board position recognition unit (5), the positional deviation amount (X_p, Y_p, θ_p) detected by the component position recognition unit (6), and the positional deviation amount (X_p, Y_p, θ_p) detected by the component position recognition unit (6). The component mounting head (8) and the X-Y table (
7) A correction value (X, Y, θ) is calculated and stored based on the drive error amount (d_x, d_y, d_θ), and the correction value (X
, Y, θ) between the component mounting head (8) and the X-Y
The correction value (X
, Y, θ) to the control unit (9).