JPH0510746A - Calibration method for surface mounter - Google Patents

Abstract

(57) [Abstract] [Purpose] To facilitate easily and accurately the calibration for obtaining the positional relationship of a circuit board or the like, which was a troublesome work in the conventional method in the surface mounter. A component supply device for feeding a component to be mounted on a circuit board to a predetermined take-out position, a component suction device for sucking the component at the take-out position, and a measurement for measuring the component sucked by the component suction device. In a surface mounter equipped with a device and a component positioning device that moves the component suction device onto the circuit board and mounts the suction component at a mounting position on the circuit board, using a reference substrate and components (chips) Carry out a calibration. In the board coordinate calibration, the component suction device is moved onto a reference component whose position is known on the reference substrate to suck the reference component, and the position is measured. From this result, the position of the reference component in the coordinate system of the component positioning device is calculated. By performing this for an appropriate number of reference components, the relationship between the coordinate system of the component positioning device and the coordinate system of the board can be obtained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a calibration method in a surface mounter for recognizing a component to be mounted on a circuit board and mounting it at a predetermined position.

[0002]

2. Description of the Related Art In recent years, there has been an increasing demand for mounting small-sized circuit parts (chips) on an electronic circuit constituting various electronic devices with higher density. Along with this demand, the component mounting technology has come to adopt the surface mounting method shown in FIG. 15 from the pin insertion method shown in FIG. That is, in the pin insertion method of FIG. 14, a large number of through holes are formed in the printed circuit board, and leads of circuit components such as resistors or capacitors and ICs are inserted into the holes from the front surface side of the circuit board so that the protruding ends of the printed circuit boards are formed. In contrast to soldering, in the surface mounting method of FIG. 15, a resistor or other circuit element or IC is installed at a predetermined position on a printed board and a lead is soldered. Therefore, it is not necessary to make holes, the leads of the circuit element are shortened, and it is suitable for high-density mounting of miniaturized parts. As a means for performing such surface mounting, a surface mounting machine (also called a chip mounter) is used.

In a surface mounter, necessary parts are sucked by a vacuum nozzle to detect the sucked state. Then, if there is a position shift or the like, it is corrected and the component is mounted on the circuit board. Therefore, the surface mounter includes a component suction device, a moving mechanism and a positioning mechanism for the component suction device, a detection device that detects the suction state of the component, and the like.

In many surface mounters, the pick-up position of a component and the mounting position of a component on a board can be directly input from data such as drawings by numerical input. In this input method, in order to calculate the coordinate value of the component mounting operation axis from the data of the take-out position and the component position data on the board,
It is necessary to know in advance the relationship between the take-out position and the coordinate system of the board and the coordinate system of the component mounting operation axis. Obtaining the relationship between the coordinate systems in this way is called calibration.

The conventional calibration method has been performed in the following procedure.

From the design drawing of the surface mounting machine,
Calculate the relationship of the ordinary coordinate system. However, since this does not take into consideration machining errors and assembly errors, it often deviates from the actual relationship.

The surface mounter is operated in accordance with the relationship obtained in step 1 to measure the error in the pick-up position and the error in the component mounting position, and the coordinate relationship is corrected based on the result.

The process is performed again according to the corrected coordinate relationship. This is repeated until satisfactory accuracy is obtained.

[0009]

However, the above method requires a device for accurately measuring the error of. On the other hand, if the error is measured by visual inspection or the like, the error is often not accurately obtained, and therefore the operation from then on is repeated many times. Further, it is necessary to perform calibration even when the mechanical positional relationship between the elements and parts of the surface mounter is deviated due to maintenance or the like. As described above, the calibration in the conventional surface mounter is a very troublesome work.

Therefore, an object of the present invention is to calibrate a surface mounter which has been a troublesome work in the conventional method.
The goal is to make it easy and accurate.

[0011]

According to the present invention, there is provided a component supply device for feeding a component to be mounted on a circuit board to a predetermined take-out position, a component suction device for sucking a component at the take-out position, and the component suction device. A surface mounter equipped with a measuring device for measuring the components sucked by the device, and a component positioning device for moving the component suction device onto the circuit board and mounting the suction component at a mounting position on the circuit board. In the calibration method in step 1, a reference board and components are prepared for obtaining the positional relationship of the circuit board, the component supply device, or the component suction device with respect to the component positioning device, and the components are mounted at predetermined positions on the reference substrate. The component suction device is moved onto the above-mentioned reference component to suck the reference component, the position of the reference component is measured by the measuring device, and the component position is determined from the measurement result. By performing the operation of calculating the position of the reference part in the coordinate system of the device for the appropriate number of reference parts mounted on the reference board, the coordinate system of the part positioning device and the circuit board, the component supply device, or the component suction device. It is characterized in that the relationship with the coordinate system of the device is obtained.

In a specific aspect of the present invention, the coordinate system of the component positioning device is used as a reference coordinate, and the position of the reference component in the reference coordinate system and the mounting position of the reference component in the coordinate system of the reference board are obtained. It is characterized in that a transformation matrix from the reference coordinate system to the coordinate system of the reference substrate is obtained based on those values.

[0013]

According to the present invention, the calibration is performed by using the reference substrate and the component (reference chip).

In board coordinate calibration, the component suction device is moved onto a reference component whose position is known on the reference substrate to suck the reference component, and the position is measured. From this result, the position of the reference component in the coordinate system of the component positioning device is calculated. By performing this for an appropriate number of reference components, the relationship between the coordinate system of the component positioning device and the coordinate system of the board can be obtained.

In the case of the calibration of the component take-out position of the component feeder, the same operation as above is performed for the component feeder whose position is known in the coordinate system of the take-out position.

[0016]

1 is an external perspective view of an embodiment of a surface mounter according to the present invention. The surface mounter indicated as a whole by 1 is housed inside the cabinet 2 and components on a base plate (table) 3 placed on a cabinet 2 having double doors 2a and 2b on the front side. It consists of components.

On the upper surface of the table 3, there is installed a robot 5 which is a component positioning device having a component mounting head 4 which constitutes a component suction device which will be described in detail later. Further, a belt conveyor 7 for moving the printed circuit board 6 serving as a circuit board, a large number of tape cartridges 8 in which various components (chips) to be mounted on each circuit board are collectively taped by the same type, and each cartridge 8 is taped. Front tape feeder 10a and rear tape feeder 10b, which are component feeding devices for respectively feeding the above components to the take-out positions on the front feeder base 9a and the rear feeder base 9b, the monitor CRT 11, the keyboard 12 operated by the operator, and the robot operation panel. 13 and the like are installed, a controller 14 and an image processing device 15, which will be described later, are installed in the cabinet 2.

The robot 5 supports the mounting head 4 and moves the mounting head 4 in a vertical direction (Z-axis direction) perpendicular to the table 3, and a Z-axis slide block 16 (FIG. 2) and the Z-axis slide block 16. From an X-axis arm 17 that moves the X-axis arm 17 in a direction parallel to the belt conveyor 7 (X-axis direction) and a Y-axis arm 18 that moves the entire X-axis arm 17 in a direction perpendicular to the belt conveyor 7 (Y-axis direction). Become.

The X-axis arm 17 slidably supports in the X-axis direction a hollow prismatic Z-axis base 19 in which the Z-axis slide block 16 is vertically movable. Z-axis base 19
A Z-axis motor 20 for moving the Z-axis slide block 16 in the Z-axis direction is mounted on the upper end of the Z-axis slide block 16, and a Z-axis motor 20 is installed in the Z-axis base 19 as indicated by a broken line in FIG. A ball screw mechanism 21 for converting the rotation of the above into the vertical movement of the Z-axis slide block 16 is housed. Further, in the X-axis arm 17, an X-axis motor 22 and a ball screw mechanism 2 for converting the rotation of the X-axis motor 22 into a movement of the Z-axis base 19 in the X-axis direction.
3 is stored. One end of the X-axis arm 17 slides along the Y-axis arm 18, and the other end slides along the horizontal axis of a gate-shaped guide member 24 installed on the table 3 in parallel with the Y-axis arm 18. It is supported freely. Further, in the Y-axis arm 18, a Y-axis motor 25 and a ball screw mechanism 26 for converting the rotation thereof into movement of the X-axis arm 17 in the Y-axis direction are housed. The above three motors 20,
An AC servomotor is used for each of 22 and 25.

Therefore, the Z-axis slide block 16 to which the mounting head 4 is attached is driven by the Z-axis motor 20 to drive the Z-axis slide block 16.
It moves in the Z-axis direction along the axis base 19, and the X-axis motor 2
2 moves in the X-axis direction along with the Z-axis base 19 along the X-axis arm 17, and further, by driving the Y-axis motor 25, along with the Z-axis base 19 and the X-axis arm 17 along the Y-axis arm Y. Move in the axial direction.

The robot 5 has an X-axis arm 17 and a Y-axis.
Positions on the X-axis, Y-axis and Z-axis are detected by the ball screw mechanisms of the shaft arm 18 and the Z-axis base 19 and the encoders built in the motors 22, 25 and 20, and are fed back to the controller 14. Has become.

Next, the mounting head 4 of the embodiment will be described.

As shown in FIGS. 2 and 3, the head portion is
An L-shaped head base 30 fixed to the lower end portion of the Z-axis slide block 16, a nozzle index mechanism 31 installed on one side thereof, and a pulse motor mounted on the opposite side of the head base 30 as a drive source thereof. 32 and 32.

The nozzle index mechanism 31 is mainly composed of a cylindrical rotary housing 33, and three kinds of suction nozzles 34, 35 and 36 are arranged radially on the outer peripheral surface thereof at equal angular intervals (120 ° in this case). Have. These nozzles have different diameters. In the embodiment, the nozzle 34 has the largest diameter, the nozzle 35 has the middle diameter, and the nozzle 36 has the smallest diameter.

Further, as shown in FIG. 4, the nozzle index mechanism 31 is attached to a positioning cylinder 37 provided on the outer peripheral surface of the rotary housing 33, and is movably mounted parallel to the cylinder 37, and protrudes from the cylinder 37. An index and nozzle selection positioning pin 40 connected to the plunger 38 via a connecting member 39, a circular rotary plate 41 adjacent to the end face of the housing 33, and a rotatably penetrating hole along the central axis of the housing 33. And a rotating shaft 42.

A rotary plate 41 is fixed to one end of the shaft 42, and a belt wheel 43 is fixed to the other end thereof.
A belt 45 is wound around a belt wheel 43 and a belt wheel 44 attached to the rotating shaft of the pulse motor 32, as shown in FIG. 3, to transmit the rotation of the pulse motor 32 to the shaft 42.

The rotating plate 41 is provided with at least one hole (three in the figure) for receiving the tip of the positioning pin 40 for rotating the housing 33. On the other hand, the head base 30 has suction nozzles (large, medium, and small) 34, 3 respectively.
Three holes 47a, 47b (FIG. 2) and 47c (not shown) into which the base ends of the positioning pins 40 are inserted are provided to fix the housing 33 when the suction holes 5 and 36 are set to the lower suction position. The head base 30 is used as a housing fixing portion.

Therefore, for example, as shown in FIG. 2, when the suction nozzle (large) 34 is set at the lower suction position,
The base end of the positioning pin 40 has a suction nozzle (large) 3
4 into the hole 47a on the head base 30 side. When switching the suction nozzle from this state, for example, when switching the suction nozzle (middle) 35 to the suction position, the positioning pin 40 is moved to the left in FIG.
The base end portion is pulled out from the hole 47a corresponding to the suction nozzle (large) 34 on the head base side, and the tip end portion of the positioning pin 40 is put into the hole 46 of the rotary plate 41 (in the case of the figure, any of three may be used). . As a result, the rotary plate 41 and the housing 3
3 is drivingly connected. That is, the rotation of the shaft 42 causes the rotation of the housing 3 via the rotary plate 41 and the positioning pin 40.
3, the housing 33 can be rotated by driving the pulse motor 32. In this case, the housing 33 rotates until the positioning pin 40 reaches the hole 47b corresponding to the suction nozzle (middle) 35 on the head base 30 side, and after the rotation is stopped, the positioning pin 40 is moved in the opposite direction to the above. The tip end portion is pulled out from the hole 46 of the rotary plate 41, and the base end portion is put into the hole 47b on the head base side. As a result, the housing 33 is fixed again to the head base side, and the suction nozzle (center) 35 is set to the suction position.

As shown in FIG. 4, each of the suction nozzles 34, 35, and 36 has a rotary shaft 49 which is rotatably installed about its center line inside a cylindrical portion extending radially outside the rotary housing 33. It is inserted and fixed at the tip of the rotary shaft 49, and communicates with an air vent (vacuum outlet) 51 formed on the side surface of the cylindrical portion through an elongated passage 50 formed at the center of the rotary shaft 49. In use, the component 52 is adsorbed to the tip of the adsorption nozzle 34 by sucking air from the air vent 51 using a vacuum pump or the like.

A bevel gear 53 is fixed to the base end portion of the rotary shaft 49 extending toward the center of the rotary housing 33, and this gear 53 meshes with a bevel gear 54 fixed to the rotary shaft 42. Therefore, by the rotation of the shaft 42 driven by the pulse motor 32,
The suction nozzles 34, 35 and 36 rotate via the bevel gears 54 and 53 and the rotating shaft 49.

The tip of each cylindrical portion extending to the outside of the rotary housing 33 is covered with a light diffusion plate 60, and each suction nozzle 34,
Reference numerals 35 and 36 penetrate the light diffusion plate 60. The light diffusing plate 60 is irradiated with light from an illuminating device necessary for detecting a component suction state by a nozzle as described later. In the embodiment, the light necessary for detecting the component suction state is sent to each suction nozzle in a direction perpendicular to each other. Therefore, the light diffusing plate 60 diffuses or reflects the light incident from one side surface downward, and therefore, as shown in the enlarged view of FIGS. 5 and 6, the periphery of the light diffusing plate 60 is Except for the light source side
It is formed as the reflection film 61 or the reflection surface 62.

The illuminating device which irradiates the light diffusion plate 60 with light includes a casing 63 having an opening on the side of the light diffusion plate 60 and a plurality of light sources (in this case, 3) arranged in the casing 63 as shown in FIG. Light emitting diode) 64 and an optical system block 65 to which the casing 63 is fixed.

The optical system block 65 is composed of an L-shaped rectangular tube-shaped housing, and plane mirrors 66 and 67 having reflecting surfaces facing inward are disposed at the lower tip end side and the corners thereof, respectively. The plane mirror 6 on the lower end side of the optical system block 65
A circular transparent plate 68 is attached to a portion located above 6 to form a light transmission window. On the other hand, the optical system block 6
A telephoto lens 69 is housed inside the part extending upward of 5, and a close-up ring 70 located above the telephoto lens 69.
Further, the CCD (charge coupled device) camera 71 constitutes a visual device for visually detecting the component adsorption state.

Therefore, the light emitted from the light source 64 of the illuminating device passes through the light diffusing plate 60 and uniformly illuminates the component 52 adsorbed to the tip of the nozzle from above. CCD camera 7
From the perspective of 1, the component 52 can be captured as a stable image via the two plane mirrors 66 and 67.

The optical system block 65 is slidably attached in the X-axis direction to the block support portion 72 fixed to the lower end of the Z-axis base 19, and the air cylinder 73 shown in FIG.
It is moved by a drive mechanism including. The position of the optical system block 65 is detected by a pair of position detectors 74 (one of which is shown in FIG. 4) arranged at both ends of the air cylinder 73. A shock absorber 75 is installed on the upper surface of the block support portion 72.

FIG. 7 shows one cycle operation of the head section. It will be described in order as follows.

The mounting head 4 is moved to a component take-out position on the feeder base 9 by the operation of the robot 5 shown in FIG. 1, and is lowered to a predetermined height. If the nozzle selected for the component 52 sent to the take-out position by the tape feeder 10 is not at the suction position, that is, the position of the suction nozzle (large) 34 in FIG. 2, the nozzle is set to the suction position. The housing 33 is rotated until it comes, and then the component 52 is attracted to the tip of the nozzle by vacuum suction.

The mounting head 4 rises and moves toward a predetermined component mounting position, that is, a position on the printed circuit board 6 carried by the belt conveyor 7. At this time, the optical system block 65 is moved (advanced) to the left in the figure, and its tip is positioned directly below the component 52 sucked by the nozzle.

The nozzle that sucks the component 52 is rotated to set the mounting angle of the component on the substrate 6. Here, the posture of the component, that is, the position and inclination of the component are measured. This measurement is achieved by sending the image of the component captured by the CCD camera 71 to the image processing device 15 for processing, and making the determination by the controller 14.

If correction is required as a result of the measurement, after moving (retracting) the optical system block 65 to the right, the mounting head 4 itself is moved on the XY plane or the nozzle that picks up the component is rotated. Thus, the mounting position or inclination of the component is corrected.

The mounting head 4 descends and the printed circuit board 6
The component is mounted on the substrate by positioning the component on the top, stopping suction of the nozzle, and opening to atmospheric pressure.

When the mounting head 4 moves up and the next component is to be mounted, it moves to the component pick-up position.

The feature of the above-mentioned operation is that the head which picks up the component carries out the image measurement and the correction operation (the above-mentioned operation and the operation) of the component on the path for transporting the component from the component take-out position to the mounting position. This is to shorten the time.

FIG. 8 shows an example of the state of the suction nozzle and the component when measuring the component. Thus, the gravity center position (ΔX, ΔY) of the attracted component and the inclination Δ of the long axis of the equivalent inertial ellipsoid
By measuring θ, data necessary for correction can be obtained.

FIG. 9 shows the hardware configuration of the controller 14.

The controller 14 is the above-mentioned monitor C.
The image processing device 15 and the communication path RS232 to which the RT 11 and the keyboard 12 are connected and the CCD camera 71 is further connected.
A PC 80 mainly connected via C, and an X-axis and Y-axis motor control circuit 83, Z in a bus expansion box 82 connected to a bus expansion board 81 attached thereto.
The axis motor control circuit 84, theta axis motor control circuit 85, the output board 86 and the input board 87, and the output signals from the respective motor control circuits 83, 84, 85, and the corresponding X axis motor 22, Y axis motor. 25, Z-axis motor 20, θ-axis pulse motor 32, X-axis servo amplifier 88, Y-axis servo amplifier 89, Z-axis servo amplifier 90 and θ-axis pulse motor driver 91.
And an interface circuit 93 that receives an output from the output board 86 and converts it into a signal for driving the conveyor motor 92.
It has and.

The output board 86 has a solenoid valve 94 which is driven by the output to operate the positioning cylinder 37, and the air cylinder 73 (for the camera shift) which is also driven by the output from the output board 86. A solenoid valve 95 for operating (FIG. 2) is connected. The input board 87 has three sensors 96 for detecting the positions of the three nozzles 34, 35, 36.
And a sensor 97 for detecting the position of the positioning pin 40 (FIG. 4) is connected.

The personal computer 80 of the controller 14 is
Responsible for controlling the entire surface mounter such as motor control, sequence control, and program change. Further, the personal computer 80 may be connected to the host computer 98 via a communication line, or may be connected to the host computer 98 via a medium such as a flexible disk 99 and the disk unit 100 for exchanging data and the like.

The motor control circuits 83, 84, 85 and the image processing device 15 connected to the personal computer 80 have independent functions, and control the three axes of the servo motor and the pulse motor and calculate the attitude of the sucked parts. Contributes to reducing the load on the personal computer 80. In particular, the image processing device 15 constitutes a measuring device that measures the component sucked by the component suction device based on the information from the CCD camera 71 of the visual device.

The monitor CRT 11 can be used for both the screen display of the personal computer 80 and the monitor screen display of the image processing device 15.

In the surface mounter of the embodiment, in order to realize the measurement and correction operations as described above, it is necessary to accurately find the positional relationship between the substrate and the tape feeder 10 with respect to the robot 5, and to find the positional relationship. Equipped with a calibration function. In addition to the necessity of operation, if this function is used when the positional relationship is displaced due to maintenance or the like, there is also an advantage that it is not necessary to change the component position data each time.

The calibration will be described in detail below.

1. First, the coordinate system is defined as follows (see FIG. 10).

"Base" (reference coordinates): Same as the motion coordinates of the robot 5. With the tip of the nozzle in the operating coordinates as the origin, the coordinate axes are set in the X-axis and Y-axis directions of the robot 5.

"Plate" (coordinates of a positioned substrate): An angle (corner) serving as a reference of the substrate is set as an origin, and respective coordinate axes are taken in the X-axis and Y-axis directions of the substrate.

"Feed 1" (coordinates of the front feeder base 9a): A coordinate system in which the take-out position of the front feeder base 9a is the origin and the take-out position direction of the front tape feeder 10a is the X axis.

"Feed2" (coordinates of the rear feeder base 9b): A coordinate system in which the take-out position of the rear feeder base 9b is the origin and the take-out position direction of the rear tape feeder 10b is the X axis.

"Nozzle" (coordinate system of suction nozzle): The center of the nozzle is the origin, and the coordinate axis is parallel to the reference coordinate system.
This coordinate system is fixed at the tip of the robot and moves with it. There are three types of nozzles (large, medium, and small), and one of them (for example, the large nozzle 34) is used as the reference for this coordinate. The coordinates of the component mounting position correcting camera screen are also set to be the same.

Here, the _homogenous transformation matrix from the reference coordinate "Base" to each coordinate is T _Plate , T _Feed1 , T _Feed2 , and T _Nozzle , respectively.

Since each coordinate has no information of height and direction and can be considered to be on the same plane, the homogeneous transformation matrix is

[0061]

[Equation 1]

It is expressed as Here, θ is the inclination of the current coordinate with respect to the reference coordinate, and (a, b) is the position of the origin of the current coordinate in the reference coordinate.

The relationship between one point (x _1, y ₁ ) in the coordinate system represented by the above-mentioned homogeneous transformation matrix T and the coordinate value (x _0, y ₀ ) in the reference coordinates is as follows. It is expressed as follows.

[0064]

[Equation 2]

The calibration is performed by each homogeneous transformation matrix T
To find the offset between _Plate , T _Feed1 , T _Feed2 and each nozzle.

2. When the nozzle offset large nozzle 34 is used as a coordinate reference, this nozzle 34
The offset amounts of the middle nozzle 35 and the small nozzle 36 with respect to are calculated by the following procedure.

(1) The origin of the camera screen is set at the center of the large nozzle 34. At this time, if the tip of the nozzle is not at the center of rotation, the position of the tip changes depending on the rotation angle. Therefore, while rotating the nozzle, the position is measured several times, and the center position is set as the origin. For example, as shown in FIG. 11, when the position of the center of gravity when the nozzle is rotated by 120 ° and measured is represented by vectors x ₁ , x ₂ , and x ₃ , the rotation center position x ₀ of the nozzle is represented as follows. .

[0068]

[Equation 3]

(2) With respect to the origin position set in (1), the offset amounts of the middle nozzle and the small nozzle are measured by an image and stored. Also at this time, the measurement is performed several times while rotating each nozzle, and the center is taken.

3. Calibration of Substrate Coordinates Here, a method of obtaining T _Plate will be described.

Since T _Plate is a coordinate serving as a reference for calculating the component mounting position, accurate calibration is performed using a reference substrate for calibration and a visual device.

T _Plate can be obtained from the coordinate value of the point whose position on the reference substrate is accurately known in the reference coordinates.

(1) Reference Board The reference board is formed to have the same size as the printed board, and holes are made at two or more known positions. A reference chip having the same size as the chip can be inserted into the hole. The reference substrate is prepared by inserting a reference chip in a hole in advance and flowing it by a belt conveyor of a surface mounter to position it at a predetermined position.

For example, as shown in FIG. 12, the reference chip 1
01 and the position of the hole 103 for mounting it (in this case, 3
A reference substrate 102 whose position) is accurately determined is prepared. The position of the hole 103 in the coordinates of the substrate (x _pi ,
y _pi ) [i = 1, ..., N (number of holes)] is known in advance.

(2) The visual measurement robot 5 is manually moved so that the nozzle comes near the position above the hole 103. Next, after lowering the nozzle to adsorb the reference chip 101, the camera 71
The center of gravity (center) position of the reference chip 101 is measured with a visual device including. The coordinate value of the hole 103 in the reference coordinates is obtained by adding the measurement position of the visual device to the origin position of the visual device (coordinate value of the robot).

Here, a vector indicating the position of the i-th hole in the board coordinates is

[0077]

[Equation 4]

Let v _{i be} the vector of the measurement position when the sucked reference chip 101 is measured by the visual device, and w _i be the origin position vector of the visual device in the coordinate system of the surface mounting operation axis at the time of measurement. Position of the reference chip 101 (position of the hole 103) in the coordinate system of the surface mount operation axis
s _i is

[0079]

S _i = w _i + v _i (5) w _i can be found by reading the coordinate value of the motion axis, and v _i is a measured value, so it can be obtained by calculating s _i at the time of measurement.

The relationship between s _i and u _i is

[0081]

## _EQU6 ## s _i = T _Plate u _i (6) However,

[0082]

[Equation 7]

Therefore,

[0084]

[Equation 8]

Then,

[0086]

[ _Mathematical formula-see original _document ] x _ri = x _pi cos θ−y _pi sin θ + a ...
(9) y _ri = x _pi sin θ + y _pi cos θ + b (10)

However, since the actually measured value includes an error, the above relationship is not always satisfied. Therefore, T _Plate is determined from the measured v _i and s _i by the method of least squares. That is, the error in each measurement

[0088]

[ _Equation 10] e _xi = x _ri −x _pi cos θ + y _pi sin θ−a (11) _ey i = y _ri −x _pi sin θ−y _pi cos θ−b (12)

[0089]

[Equation 11]

The values of θ, a and b that minimize are obtained.

Therefore, when S _E is partially differentiated by a, b and θ,

[0092]

[Equation 12]

When these values are set to 0, the minimum value is obtained.

[0094]

[Equation 13]

From the above three expressions (17), (18) and (19)

[0096]

[Equation 14]

Here, if θ = θ ₀ + Δθ (θ ₀ is a design angle and Δθ is an error), and Δθ << 1, then cos Δθ
≈ 1, sin Δθ ≈ Δθ, so

[0098]

[Equation 15]

In addition,

[0100]

[Equation 16]

FIG. 13 is a flowchart showing the above-described board coordinate calibration operation.

First, the hole number i in the coordinates of the substrate
Is set to 1 (step 1), and the reference substrate is positioned at a predetermined position (step 2).

Next, the robot 5 is manually operated to set the nozzle near the position above the reference chip placed in the i-th hole (step 3). Here, the coordinate value (the origin position of the camera 71) w _i of the robot 5 is read and stored (step 4).

Next, the z-axis is lowered, that is, the nozzle is lowered (step 5), and the reference chip is sucked (step 6).
After the z-axis is raised (step 7), the camera is moved forward (step 8), and the center of gravity (center) position v _i of the reference chip is measured and stored (step 9). After that, the camera is retracted (step 10), the suction is released (step 11), and the coordinate value s _i = w _i + v _i of the hole at the reference coordinates is calculated (step 12).

Next, i is incremented by 1 (step 13), i
Determines whether the number of holes exceeds n (step 1
4). As a result, if “No”, the operations from step 3 onward are repeated, and if “Yes”, Δθ in the equations (22) to (24).
Θ and a and b are calculated and stored (step 1
5).

Thus, the homogeneous transformation matrix T _Plate from the reference coordinates to the substrate coordinates is obtained, and the calibration of the substrate coordinates is achieved.

4. Calibration of the feeder base coordinates Regarding the coordinates of the feeder base, fix the actually used feeder (tape feeder, etc.) to the base, and pick up the actual chip components to measure the coordinates of the base coordinates above. Perform calibration in the same way.

[0108]

As described above, according to the present invention, a reference board and parts are prepared for obtaining the coordinate positional relationship of a circuit board or the like in a surface mounter, and mounted on a predetermined position on the reference board. Adhesion and posture measurement were performed on a suitable number of standard components, and the relationship between the coordinate system of the component positioning device and the coordinate system of the circuit board was determined from the measurement results. In addition to being accurate, there is no need for special equipment other than the standard board and parts for calibration, and the effect that the user of the surface mounter can easily carry out calibration work. can get.

[Brief description of drawings]

FIG. 1 is a perspective view showing a surface mounter according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of a component mounting head portion and its peripheral portion.

FIG. 3 is a partial cross-sectional plan view of a component mounting head section.

FIG. 4 is a partial vertical cross-sectional view showing the structure of the inside of the component mounting head and its peripheral portion.

FIG. 5 is a vertical cross-sectional view of a head tip portion.

6 is a partial cross-sectional bottom view taken along the line VI-VI in FIG.

FIG. 7 is an operation explanatory view of a component mounting head unit.

FIG. 8 is a diagram showing an example of a state of a suction nozzle and a component during component measurement.

FIG. 9 is a block diagram showing a hardware configuration of a controller.

FIG. 10 is a diagram showing a coordinate system for calibration of an example.

FIG. 11 is a diagram showing the center of gravity x ₁ , x ₂ , x ₃ and the center of rotation x _{0 of the} nozzle when the nozzle is rotated by 120 ° and measured.
FIG.

FIG. 12 is a diagram showing a reference chip and a reference substrate for calibration.

FIG. 13 is a flowchart showing a procedure for performing board coordinate calibration.

FIG. 14 is a diagram showing a pin insertion method of component mounting methods on a printed circuit board.

FIG. 15 is a diagram showing a surface mounting method of component mounting methods on a printed circuit board.

[Explanation of symbols]

1 ... Surface mounter, 2 ... Cabinet, 3 ... Base plate, 4 ... Mounting head, 5 ... Robot, 6 ... Printed circuit board, 7 ... Belt conveyor, 8 ... Tape cartridge, 9a, 9b ... Feeder base, 10a, 10b ... Tape feeder, 11
… CRT, 12… Keyboard, 13… Operation panel, 14
... controller, 15 ... image processing device, 16 ... Z-axis slide block, 17 ... X-axis arm, 18 ... Y-axis arm,
19 ... Z-axis base, 20 ... Z-axis motor 21, 23, 2
6 ... Ball screw mechanism, 22 ... X-axis motor, 24 ... Guide member, 25 ... Y-axis motor, 30 ... Head base, 31 ...
Nozzle index mechanism, 32 ... Pulse motor, 33 ...
Housing, 34, 35, 36 ... Suction nozzle, 37 ... Positioning cylinder, 40 ... Positioning pin, 41 ... Rotating plate, 42 ... Rotating shaft, 43, 44 ... Belt wheel, 4
6, 47, 48 ... Hole, 51 ... Air vent, 52 ... Parts,
53, 54 ... Bevel gear, 60 ... Light diffusion plate, 63 ... Casing, 64 ... Light source, 65 ... Optical system block, 66, 6
7 ... Planar mirror, 68 ... Window, 71 ... CCD camera, 73 ... Air cylinder, 80 ... Personal computer, 83, 84, 85 ... Motor control circuit, 88, 89, 90 ... Servo amplifier, 91 ... Pulse motor driver, 94, 95 ... Solenoid valve, 96, 97 ... Position sensor, 101 ... Reference chip, 102 ... Reference substrate, 103 ... Hole.

Claims (2)

[Claims] 1. A component supply device for feeding a component to be mounted on a circuit board to a predetermined take-out position, a component suction device for sucking a component at the take-out position, and a component sucked by the component suction device. In a surface mounter including a measuring device for moving the component suction device onto the circuit board and mounting the suction component at a mounting position on the circuit board, Prepare a reference board and components to determine the positional relationship between the circuit board, component supply device, or component suction device, and move the component suction device onto the reference component mounted at a predetermined position on the reference substrate. Suck the reference component, measure the position of the reference component with the measuring device, and calculate the position of the reference component in the coordinate system of the component positioning device from the measurement result. Performing the operation for a proper number of reference parts mounted on the reference board to obtain the relationship between the coordinate system of the component positioning device and the coordinate system of the circuit board, the component supply device, or the component suction device. Calibration method characterized by. 2. A coordinate system of the component positioning device is used as a reference coordinate, and a position of the reference component in the reference coordinate system and a mounting position of the reference component in the coordinate system of the reference board are obtained, and those values are set to those values. The calibration method in a surface mounter according to claim 1, wherein a conversion matrix from the reference coordinate system to the coordinate system of the reference substrate is obtained based on the calibration matrix.