JP5638978B2 - Pressure control head of mounter - Google Patents

Description

  The present invention relates to a pressure control head of a mounter device that mounts electronic components on a substrate.

As a mounter device for an electronic component, for example, according to Patent Document 1, a Z-axis motor that lowers a nozzle that sucks and holds a component to a substrate surface and a VCM (voice coil motor) that pressurizes the component toward the substrate surface, There is known a mounting head including a θ motor that rotates a nozzle in a horizontal direction to adjust a mounting angle of components.
As in Patent Document 1, in the mounting head that can control the mounting load, the nozzle tip of the mounter head placed on the XY mechanism is moved to the parts on the board to be mounted by the XY operation, and other parts around the board. The first Z-axis height that has no possibility of contact with the mounter mechanism part, the second Z-axis height immediately before the component adsorbed by the nozzle starts to contact the substrate surface, and the component contacts the substrate surface And a third Z-axis height that is pressurized by a VCM or the like.
According to Patent Document 1, the movement from the first height to the second height is performed by the Z-axis motor, and the movement from the second height to the third height is executed by the VCM.

  In addition to such a VCM, as a pressurizing means, for example, a pressurizing method using a fluid as in Patent Document 2, for example, a pressurizing method using a compression spring as in Patent Document 3, and the like are known.

JP 2006-147640 A JP-A-10-27996 JP 2007-27408 A

However, in the conventional system, in addition to the head lifting / lowering means, the pressure means as described in Patent Documents 1 and 2 are required, so that the cost is high and the structure is inevitable.
That is, in Patent Document 1, when the VCM is a pressurizing unit, the load applied to the nozzle is measured by a load detecting unit such as a load cell, and the output of the VCM is feedback-controlled so as to be a target load.
Such a VCM method has a relatively high accuracy, but an increase in the size of the VCM is inevitable when a high load is realized.
In addition, as in Patent Document 2, in the method of obtaining pressure by the air pressure controlled by an electropneumatic regulator or the like, the cost of an electropneumatic regulator or an air plunger is high, and the head size is relatively large. It is applied when controlling a wide load range with high accuracy.

Further, in the method using the spring force of Patent Document 3 as a pressure source, it is difficult to realize a highly accurate spring that covers a wide load range even at a low cost.
That is, as in Patent Document 3, in the method of obtaining the applied pressure by the compression force / tensile force of the spring, an inexpensive and compact configuration can be realized, but a spring having a wide load range such as 1N to 50N is used. Since it is difficult to achieve high accuracy, it can be applied only to applications with a relatively narrow load range.

  An object of the present invention is to provide a pressure control head with a simple configuration at a low cost in a mounter apparatus.

In order to solve the above problems, the invention described in claim 1
A mounter device comprising a servo motor that positions the height of a nozzle that sucks a component, and a pressure control head that can control a load pressing the component sucked by the nozzle against a substrate,
The servo motor is also used as a pressurizing source for the nozzle to press the component against the substrate,
The pressurization pressure is variable by the generated output torque of the servo motor generated by the difference between the command level logical coordinate and the actual coordinate of the servo motor ,
The generated output torque is adjusted by setting a gain corresponding to a set load of the servo motor .

The invention described in claim 2
A mounter device comprising a servo motor that positions the height of a nozzle that sucks a component, and a pressure control head that can control a load pressing the component sucked by the nozzle against a substrate,
The servo motor is also used as a pressurizing source for the nozzle to press the component against the substrate,
The pressurization pressure is variable by the generated output torque of the servo motor generated by the difference between the command level logical coordinate and the actual coordinate of the servo motor,
The generated output torque is variable based on a control parameter including a position feedback gain of the servo motor.

The invention according to claim 3
A pressure control head of the mounter device according to claim 2 ,
The control parameter is characterized in that the effectiveness of the integral compensation gain parameter is set low.

  According to the present invention, since the servo motor for positioning the height of the head is also used as a pressurizing source for pressurizing the nozzle, the pressurizing control head can be provided at a low cost with a simple configuration.

It is a schematic block diagram which shows the structure of one Embodiment of the mounter apparatus to which this invention is applied. It is a block diagram of the control system of a mounter apparatus. It is a block diagram of the mechanism system of a mounting head part. It is a graph which shows the relationship between the output voltage of a strain gauge, and a load value. It is the graph which showed the Z-axis motor speed with respect to the positional relationship of a component and a board | substrate, a detected load, a target load, and a Z-axis target coordinate. It is the graph which measured the relationship between a motor shaft transition angle (the difference of the target coordinate of a motor shaft, and a real coordinate) and generated torque. It is the graph which measured the change of the load which the board | substrate surface receives after the components adsorbed | sucked to the nozzle front-end | tip contact | abutted to the board | substrate surface. It is a block diagram of the mechanism system of a mounting head part. It is a graph explaining pressurization mounting operation. The graph shows the current waveform of the motor when the load is actually controlled, and is a graph when the feed amount is small. The graph shows the current waveform of the motor when the load is actually controlled, and is a graph when the feed amount is large. It is a schematic block diagram which shows the structure of the electronic component mounting apparatus as one Example of this invention. It is a block diagram of the mechanism system of a mounting head part. It is the figure which showed the time of component mounting. The figure shows the configuration of the suction nozzle, and is a diagram (a) showing a normal time (contact time) and a diagram (b) showing a rigid time. It is the graph which showed a pressurization waveform. It is the graph which showed the output waveform of a pressure detection part. It is the schematic which shows the mechanical structure required in implementation. It is the block diagram which showed the structure regarding control. It is a flowchart which shows the procedure of the preparation which controls a pressurizing force. It is a graph which shows the relationship of the pressurization value with respect to a motor drive current. It is a flowchart which shows mounting low pressure area control. It is a flowchart which shows mounting high pressurization area control. It is a graph which shows the motor gain for load control. It is a flowchart which shows an error determination processing sequence. It is a graph which shows an electric current value waveform when lowering a Z-axis at a fixed speed on the same conditions as the time of pressure mounting. It is a flowchart which shows the acquisition sequence of a cogging torque value. It is a graph which shows the result of having changed the operation gain of an axis and acquiring the same current waveform. It is a graph which shows the change of the electric current value by the state change of an axis | shaft. It is a flowchart which shows acquisition of a calculation formula.

DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings.
<Summary 1>
In the present invention, the Z-axis raising / lowering drive and pressurization control are realized by using a Z-axis motor that moves the Z-axis up and down as the sole drive source.
As a structure for realizing such a goal, as shown in Patent Document 3, it is common to provide a pressure spring and transmit a pressure corresponding to the compression of the spring to the nozzle. As a device that does not use a pressurizing spring, the servo motor (drive) is selected according to the difference between the Z-axis target height position (= servo motor command level logical coordinates) and the actual height position (= servo motor actual coordinates). The principle is that a generated output torque of a current amplification circuit incorporated in the circuit passes a current according to the position deviation and the position control gain to the motor winding is generated.
For this reason, when the nozzle tip is pressurized, there is a difference between the command level logical coordinates of the servo motor and the actual coordinates, and a phenomenon occurs in which the generated output torque of the motor increases. When the pressure value of the nozzle increases or decreases, the “difference between the servo motor command level logical coordinates and the actual coordinates” changes, and the same effect as the pressurizing structure using a spring is produced.
In other words, when a deviation between the target position and the actual position is detected in the position control, when the feedback gain is increased, the torque is driven so as to make the deviation zero, but when the gain is low, the deviation is low. Since it is driven by torque, the strength of the spring can be arbitrarily set by selecting the gain so that the generated output torque becomes an appropriate value.
Therefore, as will be described later, the gain for detecting the moment when the electronic component adsorbed to the nozzle is lowered and contacts the substrate, and the gain for pressing the electronic component on the substrate surface for a predetermined time with a predetermined pressure are appropriately selected. By doing so, desired load control can be performed.

  Thus, since the spring effect is proportional to the set position control gain, the generated output torque of the servo motor is appropriately set to the feedback gain of the servo motor position control corresponding to the set load. Since it can be arbitrarily adjusted by setting, it is possible to cope with a wide range from a low load to a high load with the same structure.

Further, in the present invention, as disclosed in Patent Document 1, there is no need to “detect that a component has contacted the board surface and switch the Z-axis drive motor from position control to torque control”.
That is, from the first height where the nozzle tip of the mounter head placed on the XY mechanism does not come into contact with components on the board to be mounted or other mounter mechanism around the board by the XY operation. In the second Z-axis height immediately before the component adsorbed by the nozzle starts to contact the substrate surface, the Z-axis is moved at a high speed in order to improve the mounting processing performance.
From the second height to the completion of component mounting including component pressure processing, the Z-axis motor is lowered at a low speed (about 4 mm / second) with the position controlled (the nozzle is advanced toward the substrate surface). When the load cell measurement result reaches the set load, the driving of the motor is stopped.
Further, until the actual coordinate of the motor reaches the target coordinate, the control is performed so as to repeat the descent unless the measured value of the load cell exceeds the set value.

  As described above, there is no need to switch the motor to torque control after the component comes into contact with the board surface, and the Z-axis motor is always operated in the position control mode, thereby managing the height coordinate for mounting the component. On the other hand, reliable pressure mounting with reference to the measurement result of the load cell can be realized.

(Embodiment 1)
<Mounter structure>
FIG. 1 is a schematic configuration diagram of an electronic component mounting apparatus (mounter apparatus).
As shown in the figure, the electronic component mounting apparatus 1 is disposed on the circuit board transport path 15 extending in the left-right direction slightly rearward from the central part, and on the front part (lower side in the figure) of the apparatus 1. A component supply unit 11 that supplies components to be mounted on the apparatus 1, and an X-axis movement mechanism 12 and a Y-axis movement mechanism 14 that are disposed at the front of the apparatus 1 are provided.

A component recognition camera (imaging unit) 16 that images the component sucked by the suction nozzle 131 from below is disposed on the side of the component supply unit 11.
The X-axis moving mechanism 12 moves a mounting head unit 13 (pressure control head) including a suction nozzle 131 that sucks components in the X-axis direction.
The mounting head unit 13 is connected to the X-axis moving mechanism 12.
The Y-axis moving mechanism 14 moves the X-axis moving mechanism 12 and the mounting head unit 13 in the Y-axis direction.
The mounting head unit 13 includes a Z-axis moving mechanism that moves the suction nozzle 131 up and down in the vertical direction (Z-axis direction), and also rotates the suction nozzle 131 about the nozzle axis (suction axis). A moving mechanism is provided.
The mounting head unit 13 is mounted with a substrate recognition camera 17 that images a substrate mark formed on the circuit substrate 10 so as to be attached to the support member.

  FIG. 2 shows the configuration of the control system of the electronic component mounting apparatus. In the figure, reference numeral 20 denotes a microcomputer (CPU) for controlling the entire apparatus, and a controller (control means) comprising RAM, ROM, etc. A display device (monitor) 31 is connected to the controller 20 from an X-axis motor 21. Each has control.

The X-axis motor 21 is a drive source for the X-axis moving mechanism 12 and moves the mounting head unit 13 in the X-axis direction.
The Y-axis motor 22 is a drive source for the Y-axis moving mechanism 14 and drives the X-axis moving mechanism 12 in the Y-axis direction, so that the mounting head unit 13 can move in the X-axis direction and the Y-axis direction. .

The Z-axis motor 23 is a drive source of a Z-axis drive mechanism (not shown) that moves the suction nozzle 131 up and down, and moves the suction nozzle 131 up and down in the Z-axis direction (height direction).
The θ-axis motor 24 is a drive source for a θ-axis rotation mechanism (not shown) of the suction nozzle 131, and rotates the suction nozzle 131 about its nozzle central axis (suction axis).

The image recognition device 27 performs image recognition of the component 18 sucked by the suction nozzle 131, and includes an A / D converter 271, a memory 272, and a CPU 273.
The analog image signal output from the component recognition camera 16 that images the picked-up component 18 is converted into a digital signal by the A / D converter 271 and stored in the memory 272, and the CPU 273 is based on the image data. Recognize the sucked parts.
That is, the image recognition device 27 calculates the component center and the suction angle, and recognizes the suction posture of the component.
In addition, the image recognition device 27 calculates the substrate mark position by processing the image of the substrate mark captured by the substrate recognition camera 17.
Further, the image recognition device 27 processes the image data of the component 18 imaged by the component recognition camera 16 and the board mark data imaged by the board recognition camera 17, and transfers both correction data to the control means 20.

The keyboard 28 and mouse 29 are used for inputting data such as component data.
The storage device 30 is configured by a flash memory or the like, and is used to store component data input by the keyboard 28 and the mouse 29, component data supplied from a host computer (not shown), and the like.
The display device (monitor) 31 displays component data, calculation data, an image of the component 18 captured by the component recognition camera 16, and the like on the display surface 311.

Actually, at the stage where the production of the board is started and the components are mounted on the circuit board, the board correction data (Δx, Δy, Δθ) of the circuit board 10 based on the board mark imaged in advance by the board recognition camera 17 is stored in the storage device 30. Stored in
Then, the component supplied from the component supply apparatus 11 is sucked by the suction nozzle 131, the mounting head unit 13 is moved to the upper part of the component recognition camera 16, and the component is imaged by the camera.
The captured image of the part is subjected to image processing by the image recognition device 27 and the correction data is transferred to the control means 20.
The control means 20 reads the board correction data and the component data of the component from the storage device 30, and based on the component data and the component center and component inclination calculated by the transferred image recognition device 27, the component Recognize the mounting position and suction posture of the.
Subsequently, there is a positional shift between the component mounting position, the component center, and the suction center, and when an angular shift is detected, these total positional shift and angular shift are the X-axis motor 21, Y-axis motor 22, and θ-axis. The correction is made by driving the motor 24, and the component is mounted in a correct posture (reference angle) at a predetermined circuit board position.

  Next, the mounting head unit 13 will be described with reference to FIG.

As shown in the figure, a linear guide 101 is installed on the base frame 100 of the mounting head unit 13 so that the vertical Z driving unit 102 can move in the vertical Z-axis direction.
A Z-axis motor 23 for moving the vertical Z drive unit 102 vertically up and down is fixed to the base frame 100 at the top of the mounting head unit 13, and a screw part 111 of a ball screw is coupled to the Z-axis motor 23 via a coupling 110. Is connected.

A θ-axis motor 24 for rotating the components includes a vertical rotation drive unit bearing 105 that includes a spline bearing 107 and a rotation bearing 106 and has a belt pulley attached to the outer periphery, a θ motor pulley 108, and a timing belt 109. Connected through.
The vertical rotation drive unit bearing 105 has a spline bearing 107 therein and is connected to a nozzle shaft 104 which is a spline shaft.
A rotation bearing 106 is attached to the outer periphery of the vertical rotation drive unit bearing 105. The outer periphery of the rotary bearing 106 is fixed to the base frame 100, and the nozzle shaft 104 is fixed by the vertical rotation drive unit bearing 105 so that the rotary operation and the vertical operation can be performed.

At one end of the vertical Z driving portion 102, a nut portion 118 that meshes with the screw portion 111 of the ball screw is fixed.
Therefore, by rotating the Z-axis motor 23, the vertical Z driving unit 102 is driven up and down by the nut portion 118 of the ball screw, and the nozzle shaft 104 and the suction nozzle 131 can be driven up and down.

In addition, the vertical Z drive unit 102 is provided with a lower rotary bearing 141 and an upper rotary bearing 142 for rotationally supporting the nozzle shaft 104.
A deforming portion 112 having a circular hole shape is provided between the nozzle shaft 104 of the vertical Z driving portion 102 and the nut portion 109 of the ball screw. A strain gauge 113 is attached to the deformed portion 112.
In the strain gauge 113, the relationship between the output voltage of the strain gauge 113 and the load value is preliminarily calibrated and takes the relationship as shown in FIG. 4 and stored in the controller 20.
The strain gauge 113 can be replaced with a load cell with an appropriate structural change.

  In addition, an origin sensor 114 is fixed to the base frame 100 so as to detect the vicinity of the fixed portion of the vertical Z driving unit 102 on the linear guide 101 side.

  Next, the flow of the pressure mounting operation of the electronic component will be described.

The X-axis moving mechanism 12 and the Y-axis moving mechanism 14 of the mounting head 13 shown in FIG. 1 are operated to move the mounting head unit 13 above the electronic component supply device 11 and suck the electronic component 18.
The mounting head portion 13 that has attracted the electronic component 18 is moved above the component recognition camera 16 to recognize the electronic component 18.
After the recognition is completed, the mounting head unit 13 is moved, and the electronic component 18 is adsorbed by the mounting head unit 13 to the portion where the electronic component 18 is to be mounted on the circuit board 10 and moved onto the component recognition camera 16. 18 is recognized on the component recognition camera 16 and moved to the mounting position on the circuit board 10 for mounting.

  Next, a component mounting operation by load control will be described.

The Z-axis motor 23 is driven at the mounting position of the mounting head unit 13 on the circuit board 10, and the vertical Z driving unit 102 and the suction nozzle 131 are lowered.
The component 18 sucked by the suction nozzle 131 is lowered at a high speed to the position (Z1) immediately before the mounting height on the circuit board 10 to be mounted.
Thereafter, the Z-axis motor 23 is driven to lower the component 18 sucked by the suction nozzle 131 at a low speed of about 4 mm / second to lower the target load height while suppressing the impact load.
When the lower surface of the electronic component 18 comes into contact (0) with the circuit board 10, the detection unit 112 is deformed, and a change (L0 → L1) occurs in the output of the strain gauge 113.
Further, the Z-axis motor 23 is driven so that the output of the strain gauge 113 becomes a target pressurization amount.
When the output of the strain gauge 113 reaches an output (L2) corresponding to the target load value, the Z-axis motor 23 is stopped.
After the electronic component 18 is pressure-mounted, the vacuum air is turned off, the Z-axis motor 23 is operated, and the vertical Z-axis drive unit 102 and the suction nozzle 131 are raised.
Thereafter, the next electronic component is moved to the suction position.

  Next, the origin return operation will be described.

The Z-axis drive unit 102 is moved downward (for example, 2 mm) away from the detection range of the origin sensor 114.
Then, it is raised at the origin return speed of 10 mm / second.
The detection ON height (A0) of the origin sensor 114 is read from the encoder value of the Z-axis motor and stored in the CPU 27C.
By setting the encoder origin position of the Z-axis motor 23 detected immediately after the Z-axis drive unit 102 detects the origin sensor 114 as the origin of the Z-axis drive unit 102, the Z-axis can be repeatedly turned off and on. High reproducibility of the origin can be obtained.

<Description of operation>
In FIG. 5, “1” indicates that there is no possibility that the tip of the nozzle of the mounter head placed on the XY mechanism will come into contact with components on the board to be mounted or other mounter mechanism around the board by the XY operation. 1 is a second Z-axis height immediately before the component adsorbed by the nozzle starts to contact the substrate surface.

  From the first height “1” to the second height “2”, the vehicle descends at a high speed (about 900 mm / second) in order to minimize the reduction in mounting performance. At this time, the gain is set high in order to perform normal position control in order to realize the lowered position control of the nozzle with high accuracy.

At the second height “2”, the motor shaft is stopped once, the gain is set low, and then the shaft lowering is resumed.
Here, the motor speed is set to a low speed (4 mm / second), and the gain for generating the generated torque in the motor position control mode is set according to the required set load. The generated output torque is varied based on a control parameter including a position feedback gain of the servo motor.
In addition, the setting characteristic as a gain control parameter is changed to a gain with a small number of integral compensation gain parameters.
The generated torque of the servo motor is determined according to the position feedback gain and the position deviation, but since the feedback adjustment of the generated torque functions according to the position deviation amount and the duration of the state, the time axis such as the integral compensation gain is set. The feedback function as an element is set to be small and a stable pressure is obtained.

  In FIG. 5, “3” is a contact point between the lower surface of the component and the substrate surface.

  As long as the load cell measurement result does not reach the specified load value, the Z-axis is continuously lowered, so if the component adsorbed on the tip of the nozzle comes into contact with the board surface and is prevented from descending, the target coordinate of the motor axis Since the deviation of the actual coordinates advances, the torque generated by the servo motor increases as a result of the gain setting described above, and the measurement result of the load cell also increases temporarily as the target coordinates of the motor axis advance.

FIG. 6 is a graph obtained by actually measuring the relationship between the motor shaft transition angle (the difference between the target coordinate and the actual coordinate of the motor shaft) and the generated torque. The horizontal axis indicates the difference in the coordinates as a rotation angle (10 degrees to 60 degrees) of the motor shaft.
For the generated torque, the set gain 10 to the set gain 100 are evaluated for each numerical value 10.
As described above, the motor shaft shift angle, the gain, and the motor output torque are generally in a proportional relationship. Therefore, as described above, the generated output torque can be adjusted by selecting the feedback gain of the servo motor in response to the set load (the pressing force of the component on the board surface that varies depending on the component type).

FIG. 7 shows actual changes in the load applied to the substrate surface after the component adsorbed on the nozzle tip contacts the substrate surface (Z-axis speed = 10 mm / second, head mass = 250 g).
Immediately after the component comes into contact with the board surface, the load shown on the vertical axis rises steeply, and thereafter the load increases in proportion to the transition of time.
The load portion that rises steeply is obtained by measuring the impact load due to the mass of the head movable portion.

In FIG. 5, as indicated by “4-1”, when the measurement result of the load cell exceeds the set load, the driving of the Z-axis motor is stopped.
If the minimum pressurization duration is set by the operation setting of the mounting machine, the Z-axis motor may be repeatedly lowered and raised according to the measurement result of the load cell in order to maintain the specified pressure up to “5”. It is possible.
It is desirable to stop the descent of the Z-axis motor instantaneously when the load cell measurement result exceeds the set load.
If there is a delay in this reaction, the target coordinates of the motor will advance and the generated torque of the motor shaft will be excessive.

  In the present invention, by using the relationship between the motor shaft transition angle, the gain value, and the motor output torque, the gain is set to be low, so that even if a difference between the target coordinate of the motor and the actual coordinate occurs, an extremely large load change occurs. Since it does not occur, pressure control along the target load can be realized even if the target coordinate of the motor advances too much.

Here, when the gain is appropriately set so that the maximum delay with respect to the target coordinate at the nozzle tip at the set load is 0.75 mm, the control error for the target load is 3% and the speed of the Z-axis motor is 5 mm. Assuming / second, the reaction delay allowable time = (maximum delay amount × control accuracy) ÷ motor shaft speed = 0.0045 seconds.
For this reason, in the load system of the present invention, it is possible to realize load control by controlling a normal servo amplifier directly from the mounter control unit.
Therefore, it is not necessary to provide a special control system that performs high-speed feedback control on the motor shaft, VCM, or electropneumatic regulator by referring to the measurement result of the load cell for load control.

  As described above, according to the pressure control head of the mounter apparatus of the embodiment, the effects listed below can be exhibited.

1) The head is moved from the second height position to the third height position by a Z-axis motor provided as means for moving the head from the first height position to the second height position, and the nozzle is pressurized. Since the same Z-axis motor is used as a pressurizing source, the pressurizing control head can be provided at a low cost with a simple configuration.

2) Instead of a spring with a wide load range required for pressurization, it can be pressurized with the torque generated by the motor shaft, so the structure is simple and the accuracy over a wider load range Can cope well.

3) Since the Z-axis motor is always used in the position control mode, no error occurs in the motor axis coordinate management.

4) Since the pressurization is performed using the torque generated by the Z-axis motor, the maximum deviation amount between the target coordinates and the actual coordinates of the motor can be limited.
In other words, the spring-pressurized structure has characteristics such that the load decreases suddenly immediately after passing through the load load peak (passage through the return prevention part becomes the maximum load) as in the connector insertion process. If the load has a load, an overshoot of the shaft corresponding to the amount of deflection of the spring occurs, and there is a risk of damage to the parts due to the impact load.
On the other hand, in this method, since the maximum deviation amount is limited, the position overshoot with respect to the load fluctuation is small, and the impact load can be kept small.

5) A dedicated servo amplifier with a load control function is not required.

<Outline of Invention 2>
The present invention provides a minimum displacement that the shaft moves when the operation of maintaining the target pressurization amount is maintained in the mounting head that is mounted while taking out the component from the component supply unit and controlling the load on the substrate. The load is controlled while moving minutely with the amount as a unit of movement.

Japanese Patent Application Laid-Open No. 06-177179 discloses a chip component mounting apparatus for mounting a chip component on a mount portion of a base, a chuck means for holding the chip component, a linear motor for moving the chuck means up and down, and the chuck And a control means for gradually increasing the torque of the linear motor and gradually increasing the load on the chip part when the chip part held by the means is placed on the mounting portion of the base and pressed. A mounting device is proposed.
This chip component mounting apparatus gradually increases the torque of the linear motor that drives the chuck means when the chip part held by the chuck means is placed on the mounting portion of the base and pressurizes it, and the load on the chip part is increased. Increase gradually.

However, the above-mentioned Japanese Patent Laid-Open No. 06-177179 has a problem that when the motor torque is very small, the shaft hardly operates and no change in pressure is observed.
In other words, when the load fluctuation is small during load control and the output motor torque is very small, there is an effect of mechanical sliding resistance, backlash, etc. This means that no change is seen, and no change in load is seen. For this reason, there is a problem that leads to a shift in load value or an increase in variation.

(Embodiment 2)
In FIG. 8, as in the first embodiment described above, 13 is a mounting head unit, 23 is a Z-axis motor, 24 is a θ-axis motor, 100 is a base frame, 101 is a linear guide, 102 is a vertical Z drive unit, and 104 is a nozzle. A shaft, 109 is a timing belt, 110 is a coupling, 111 is a thread portion of a ball screw, 118 is a nut portion of the ball screw, and 131 is a suction nozzle.

  Next, the pressure mounting operation will be described with reference to FIG.

<Move to pressurization start height>
“1” First stage descent: Descent to a target position at a specified speed (default: maximum speed).

<Pressurizing + pushing>
“2” gain switching: When the load control switching height is reached (in this case, contact), the shaft gain is switched to the load control gain, and the shaft speed is switched to the first speed.

“3” Feed amount switching: When the specified load is reached, the shaft operation is switched to intermittent operation and the descent starts (load cell value reading 1 [msec] interval), and the shaft feed amount is changed to a minute feed of 20 [μm]. Switch.
The intermittent operation referred to here is a repeated operation of 20 [μm] movement → stop and load cell value reading → 20 [μm] movement.

“4” Stop: When the load cell value reaches the specified load, the intermittent operation is stopped.

<Pressurization + time>
“5” maintenance: Monitors the motor current value and maintains the specified load during the specified pressurization time.

<Installation completed>
“6” Completion: The gain is switched to the gain for normal operation, and the ascending operation is performed.

10 and 11 show current waveforms of the motor when the load is actually controlled.
When the feed amount is small, the waveform is as shown in FIG. 10, and the variation (3σ) when the same load is repeatedly applied is ± 1.34 [N].
As shown in FIG. 16, by increasing the feed amount, the variation (3σ) when the same load is repeatedly applied (3σ) = ± 0.15 [N] can be reduced to about 1/10.

As described above, according to the present invention, it is possible to suppress deviations and variations in load values.
That is, since the minimum feed amount at which the motor shaft always operates is used, only the target coordinates of the motor change and the nozzle tip does not move, so that load control can be performed reliably.

(Other examples)
In the third embodiment, 20 [μm] feeding is performed at 1 [msec] intervals, but the actual change in the load at the nozzle tip can be further reduced by further reducing the time interval and feeding at high frequency.
The feed amount and the time interval are determined by the configuration of the apparatus, and are not determined by the values in the embodiment.

  In the third embodiment, the load control based on the detection of the load cell and the current control are mixed. However, the load control is effective with a mechanism having many sliding elements downstream from the load detection device. It may be effective only by control.

<Summary 3>
The present invention has a stopper portion that regulates the movement of the nozzle slider portion that moves up and down within the component suction nozzle, and has an elastic body between the nozzle and the nozzle slider, and the elastic body has a pushing amount of the component suction nozzle. In an electronic component mounting apparatus that mounts an electronic component on a substrate that is configured so that the load changes according to the pressure, when the amount of pressurization is low, pressure is applied by an elastic body, and when the amount of pressurization is high The nozzle is in contact with the stopper and pressurizes in a rigid state.

Japanese Unexamined Patent Application Publication No. 2009-277850 proposes an electronic component mounting apparatus and mounting method.
In this apparatus, the rotational force of the drive source is transmitted to the ball screw, thereby performing a minute operation in the vertical direction of the slider portion. When electronic parts are mounted under pressure, the rotational force of the pressurizing drive source (servo motor) placed at the top of the slider is transmitted to the pressurizing tool through a linear guide, and the slider is pressed by pushing it toward the board. I do. The slider part is always pulled upward by a compression spring, and the slider part is separated from the vertical drive source and the pressurizing drive source, so that it operates by separating the fine movement in the vertical direction from the pressurizing action. be able to.

However, the above-mentioned Japanese Unexamined Patent Publication No. 2009-277850 has the following three problems.
1) Impact load increases When the unit for pressurization is completely rigid (rigid body), it is necessary to suppress the descending speed of the unit as much as possible in order to alleviate the impact at the time of contact with the substrate. , Mounting time will be longer.
2) The arrival time to the target load becomes long. In order to suppress the impact load, the lowering speed is suppressed, so that the arrival time to the target load becomes long.
3) Difficult to control minute load When the unit for pressurization is completely rigid (rigid body), the pressurization amount (displacement amount) is large, so the minute load control is difficult, and the deviation from the target load value. Occurs or the variation becomes large.

(Embodiment 3)
FIG. 12 shows the configuration of an electronic component mounting apparatus according to an embodiment of the present invention. The component mounting head 201, an X-axis frame 202 driven by a motor combined with a ball screw or belt, a linear motor or the like, a Y-frame axis 1 shows an electronic component mounting apparatus that includes a substrate transfer unit 204, a substrate transfer unit 204, and a component supply unit 205.

The X-axis frame 202 is driven by motors attached to the left and right Y-axes, YL and YR axes.
The component mounting head 201 installed on the X-axis frame 202 is an XY unit that moves along the X-axis frame 202, and the X-axis frame 202 moves along the orthogonal Y-axis frame 3.
The component mounting head 201 is equipped with a nozzle for sucking a mounted component that can move in the vertical direction. The component supply unit 205 supplies the nozzle to the substrate that is transported and fixed by the substrate transport unit 204. The component can be mounted on the substrate after vacuum suction.

  Next, the configuration of the pressure head is shown in FIGS.

  The electronic component 15 is sucked, mounted, and pressurized as follows.

When the Z motor 216 attached to the fixed bracket 207 rotates, the ball screw 221 directly connected by the coupling 217 rotates, and the slider portion 223 supported by the ball screw 221 moves up and down.
The slider portion 223 has a guide nut 210a, and can be smoothly moved up and down by being connected to a linear guide rail 210b attached to the fixed bracket 207.
Further, the vertical direction of the spline shaft 206, the coupling 209, the nozzle shaft 212, and the nozzle 214 is held by the bearing in the hollow load cell 222 of the slider portion 223, and the electronic component 215 is moved up and down together with the slider portion 223. Adsorption and mounting.
Reference numeral 213 denotes a ball bush, and 219 denotes a thrust bearing.

  Next, the rotation operation of the electronic component 215 is performed as follows.

When the θ motor 220 attached to the fixed bracket 207 for rotating is rotated, the spline nut 208 connected by the timing belt 218 is rotated, and the spline shaft 206 is rotated.
Since the spline shaft 206 and the nozzle shaft 212 are directly connected by the coupling 209, when the spline shaft 206 rotates, the nozzle shaft 212 rotates and the electronic component 15 rotates.

  Next, the component suction nozzle 214 is shown in FIG.

FIG. 15A shows the nozzle 214 at the moment when the electronic component 215 and the substrate 224 come into contact with each other. FIG. 15B shows that the nozzle spring 228 is compressed by pushing the nozzle 214 downward, and the lower end of the nozzle outer 226 is compressed. And the nozzle slider stopper 225b are in contact with each other, and the nozzle 214 is shown in a rigid state.
By shortening the stroke 229 until the nozzle changes from the normal state to the rigid state as much as possible, the arrival time to the target load of the nozzle spring 228 pressure or more can be shortened.

  With the above configuration, the load detection operation when the electronic component 215 is mounted will be described with reference to FIGS.

  The force received when the nozzle 214 holding the electronic component 215 comes into contact with the substrate 224 passes through the nozzle shaft 212 and is transmitted to the hollow load cell 222 via the stepped portion of the nozzle shaft 212 to detect the amount of pressurization.

In FIG. 15A, the amount of pressure applied to the substrate in the state of this figure is the initial pressure F0 of the nozzle spring 228 (see FIG. 16). When the nozzle 214 is pushed downward from that state, the amount of pressurization increases in proportion to the spring constant of the nozzle spring 228.
That is, the amount of pressurization using the nozzle spring 228 is F1 at the maximum, and this is the amount of pressurization at which the nozzle stroke 229 shown in FIG. The pressurization amount F2 in FIG. 16 indicates that the target load value is during the compression of the nozzle spring 228.

In the state of FIG. 15B, the nozzle 214 is directly connected to the nozzle shaft 212, the hollow load cell 222, the ball screw 221, and the Z motor 216, and the torque of the Z motor 216 is transmitted to the substrate at 1: 1.
The pressurization amount F3 in FIG. 16 is a pressurization amount larger than the pressurization amount F1 at which the nozzle 214 is rigid, and there are two inclinations of the pressurization waveform until the pressurization amount F3 is reached.

  The inclination of the waveform of the pressurization amount when the nozzle spring 228 is compressed is represented by θ1 in FIG. 16, the inclination of the pressurization amount when the nozzle 214 is rigid is represented by θ2, and θ1 <θ2 is established. The pushing operation in the rigid state 214 is to transmit the torque of the Z motor 216 to the substrate with a force of 1: 1 as described above because there is no buffer material such as the nozzle spring 228.

As described above, according to the present invention, the following effects can be obtained.
1) Mitigating the impact By reducing the initial load of the nozzle spring, it is possible to reduce the impact load when contacting the substrate, and it is not necessary to suppress the descending speed, and the target load is reached in an early time. can do.
2) Can pressurize to the target load at high speed When the pressure is low, the nozzle spring is compressed and reaches the target pressure before it hits the stopper, so the time to reach the target pressure is short.
Further, by making the stroke of the nozzle spring as small as possible, the nozzle becomes rigid, and the arrival time to the high load region can be shortened.
3) Highly accurate control is possible when the amount of pressure is low. When the amount of pressure is low, the change in the amount of pressurization with respect to the amount of displacement is small by the nozzle spring.

  Further, as in the embodiment, the high-load nozzle is provided with a wide pedestal for pressurization, and can perform high-speed suction (without using load control) due to the effect of the nozzle spring.

(Other examples)
In the third embodiment, a method in which a load cell for load detection is incorporated in the Z slider portion is used. However, the same effect can be recognized in torque control in which load detection is detected by the load torque of the Z motor.

  In the third embodiment, the control is based on the detection by the load cell, but current control is also possible.

<Summary 4>
The present invention is a mounter device that moves and mounts an electronic component onto a substrate from a component supply unit using a suction nozzle, and has a sensor that detects the pressure with which the suction nozzle presses the component against the substrate surface, and the pressing force of the nozzle is It is a structure obtained from the thrust of a dedicated pressurizing motor or nozzle raising / lowering means, and controls the pressurization with reference to the output of the pressure sensor within a range of less than 50% of the set pressure range, The portion exceeding the pressurizing range is a system in which pressurization is controlled with reference to the current value of the motor.
Furthermore, utilizing the fact that the output value of the pressure sensor obtained in a range of less than 50% of the set pressure range and the current value of the motor are in a proportional relationship, the portion exceeding the pressurizing range is the current of the motor. The pressure is controlled by estimating the pressure from the value.

In Japanese Patent No. 2877120, low pressure and high pressure are prepared for the pressure detection unit of the electronic component mounting apparatus that controls the pressure when mounting the electronic component by the pressure value detected by the pressure detection unit. When the set pressure value is exceeded, a technique for switching from a low pressure detection unit to a high pressure detection unit is disclosed.
The merit is that the pressure detection unit for high pressurization can solve the problem that it cannot be read well due to the influence of noise or the like at low pressurization.

Next, the merit is supplemented.
FIG. 17 shows an example of an output when there is no load in an electrically good environment of the pressure detector.

  This pressurization detection unit was set so that the rating of the applied pressure was 100 N and 10 V was output at that time. The noise component at this time was ± 10 mV.

  Thus, since it is common for noise to be added to the output of the pressure detection unit, if a noise of ± 10 mV is applied, the output of the pressure detection unit is 10 N at 10 N, for example. Since / 100N × 10V = 1V, an error of ± 1% is detected. At 1N, an error of ± 10% is detected.

However, the above-mentioned Japanese Patent No. 2877120 uses a pressurization detection unit for low pressurization that outputs 10 V at a pressurization rating of 10 N and a pressurization detection unit for high pressurization that outputs 10 V at a pressurization rating of 100 N. By using the method of No. 1, the error due to noise becomes ± 10% even at a low pressure of 1N, and the problem can be alleviated.
However, the pressure detection unit is generally expensive, and the method disclosed in Japanese Patent No. 2877120 has a demerit that two or more must be used.

(Embodiment 4)
FIG. 18 shows a mechanical configuration necessary for implementation. In the figure, as in the first embodiment, 13 is a mounting head unit, 18 is an electronic component, 23 is a Z-axis motor, 24 is a θ-axis motor, 100 is a base frame, 101 is a linear guide, and 102 is a vertical Z drive unit. , 104 is a nozzle shaft, 105 is a vertical rotation drive unit bearing, 106 is a rotation bearing, 107 is a spline bearing, 108 is a θ motor pulley, 109 is a timing belt, 110 is a coupling, 111 is a thread portion of a ball screw, and 118 is a ball A nut portion of the screw, 131 is a suction nozzle, 143 is a rotary bush bearing, 144 is a pressure detection unit, and 150 is a pressurizing base.

  In this embodiment, a case where the control range of the pressure applied to the electronic component is 0 to 100 N will be described.

FIG. 18 shows that the pressure applied to the electronic component is transmitted to the pressure detection unit with as little loss as possible in order to reduce the pressure detection error due to sliding or the like.
In this configuration, the force that pushes the suction nozzle 131 vertically upward is detected by the pressure detection unit 144 provided on the nozzle shaft 104.
The Z-axis motor 23 used in this embodiment is a servo motor.

  Next, FIG. 19 shows a block configuration relating to the control required in the present invention. In the figure, 123 is a motor encoder, 124 is a servo driver, and 125 is a CPU.

As shown in the figure, the CPU 125 sends a position command and pressurization to the servo driver 124 based on pressurization information (pressurization value) from the pressure detection unit 144 and motor current information (motor current value) from the servo driver 124. Issue a command (command value).
Based on the command, the servo driver 124 outputs a motor drive current to the Z-axis motor (servo motor) 23. Motor position information is taken into the servo driver 124 from the motor encoder 123.

  Furthermore, in the present embodiment, it is assumed that the low pressurization region referred to in the above-mentioned Japanese Patent No. 2877120 is 0N to 10N, and the high pressurization region is 10N to 100N.

  Next, FIG. 20 shows a preparation procedure for controlling the applied pressure using the control block of FIG.

In the flow of FIG. 20, first, the pressure value when the nozzle 131 is not pressurized is set to 0N by calibration of the pressure detection unit, and the motor drive current I ₀ at this time is stored (step S21). .
Next, the nozzle 131 is pressed against the pressurization table 150 of FIG. 18, the motor 23 is driven to gradually lower the Z-axis (step S22), and the pressure detection unit 144 detects 10N (step S23). The motor driving current I ₁₀ at the time is stored (step S24).
In general, the servo motor current and the applied pressure are proportional to each other unless the motor speed changes, so the applied pressure from 10N to 100N can be estimated from the current values at 0N and 10N. Is calculated by the CPU 125 and stored.
This relationship is shown in FIG.

Further, when the electronic component 18 is mounted, if the pressurization command is in a low pressurization region of 0N to 10N, the pressurization value from the pressure detection unit 144 is used.
And in the high pressurization area | region where a pressurization command exceeds 10N, the pressure detection part 144 is not used, but the pressurization value estimated from the characteristic and motor drive current of FIG. 21 is used.
The procedure at this time is shown in FIG.

As shown in the figure, the command value is set and the value is stored when the mounting is started (step S31).
Next, is the stored command value in the low pressure range? (Step S32), if it is a low pressure region, the process proceeds to the next step S33, and if it is not a low pressure region, the process proceeds to step S36 in FIG.
In step S33, the electronic component is attracted to the nozzle, and the head is moved in the XY directions to the mounting position.
Subsequently, the electronic component is lowered at a low speed toward the board on which the electronic component is mounted (step S34).
Next, did the pressurization value exceed the command value? (Step S35), if the command value is exceeded, the process is terminated. If the command value is not exceeded, the process returns to step S34 and the subsequent processing is repeated.

As shown in FIG. 23, in step S36, the unit of the command value is converted from pressure to current using the characteristics shown in FIG. 21, and stored.
Subsequently, the electronic component is attracted to the nozzle, and the head is moved in the XY direction to the mounting position (step S37).
Subsequently, the electronic component is lowered at a low speed toward the board on which the electronic component is mounted (step S38).
Next, did the motor current value exceed the command value? (Step S39), if the command value is exceeded, the process is terminated. If the command value is not exceeded, the process returns to step S38 and the subsequent processing is repeated.

  As described above, according to the present invention, the effect of the Japanese Patent No. 2877120 can be achieved with a single pressure detection unit.

<Outline of Invention 5>
In the pressurization control system according to the first aspect of the invention described above, the present invention is based on the nozzle tip coordinates, the load value measured at the load cell, and the type of part in addition to the abnormal state determination result during pressurization control. The corresponding actions are separated.
Furthermore, as a means to suppress the impact load that the component receives from the board within the allowable range, the impact load calculation result calculated based on the impact time specified for each component type and the moving speed in the Z-axis direction when the component is mounted is The maximum movement speed in the Z direction is calculated within a range not exceeding the allowable impact load.

(Embodiment 5)
In this example, in FIG. 7 of the above-described first embodiment, after the pressurization time has elapsed, the vacuum air is turned off and the motor gain changed for load control is returned to the original value. After the Z-axis is moved to the coordinates where there is no more, the gain is restored, the ascending operation is performed at a high speed, and the operation moves to the next component suction operation.

<Motor gain for load control>
As described above, during the load operation, the gain is set such that the maximum delay amount with respect to the target coordinates of the nozzle tip is 0.75 mm. This is for controlling the nozzle tip coordinates at the time of load control with high accuracy by setting a gain that makes the maximum delay amount for each designated load constant as shown in FIG.
The load control gain includes a method of obtaining parameter values such as a proportional element, a differential element, and an integral element by calculation based on a designated load, and a method of obtaining from a predetermined table.

<About load operation speed>
In the above embodiment, the shaft operation speed during the load control operation is low (about 4 mm / s). However, since slowing the shaft operation speed affects the mounting tact, it is as fast as possible in terms of performance improvement. Operation is desirable.
The following two points can be cited as conditions for determining the operation during load control.
1) The impact (substrate contact) load must be below the specified load.
2) The load change amount of the reaction delay allowable time must not exceed the control accuracy.
Therefore, as a preliminary operation of load control, an actually mounted component is sucked, a load operation is performed on the substrate, and a contact load is measured to obtain an optimal descending speed satisfying the above two points.

  The contact load is calculated from impulse calculation.

Therefore, the descent speed can be calculated by calculation without performing preliminary operation.
In this case, the descending speed after contact approximates to 0 mm / s (stop for a moment) in order to calculate the contact load with the maximum value, and the collision time is the time for the data such as component type, component shape, board thickness, etc. The optimum collision time is adopted from the production program information and the contact load is calculated.
When the descending speed is calculated, the contact load during the production operation is fed back and the descending speed is finely adjusted in order to correct the actual error of the parts.

<Error judgment during load operation>
The purpose of controlling the mounting load varies depending on the mounted component, and therefore, a wide pressing range and pressing accuracy are required.

There are the following cases for the purpose of controlling the load.
1) Since it is a weak part such as a wafer die and the part is damaged when an excessive load is applied, the load value applied to the part is desired to be controlled to a certain value or less.
2) In bump parts, the bump size (height) is not uniform, and all the bumps may not be grounded correctly when mounted, so press with a constant force to make the bump height uniform. Want to.
3) I want to press fit connector parts.
In such an operation with different purposes, it is desirable to switch the determination condition of the abnormal operation and the response at the time of the abnormal operation.

  Therefore, control is performed to switch the error determination condition from the component type of the mounted component, the component outer shape, and the like.

  An error determination processing sequence is shown in FIG.

1) Start of lowering the Z-axis to the second height Lowers to the second Z-height immediately before starting to contact the “2” substrate surface in FIG. 5 of Embodiment 1 (step S41).
2) Load value acquisition The load value of the load cell is acquired before switching the motor speed and gain at the second Z height (step S42).
3) Load value determination When a load is applied to the load cell at the second Z height, an error occurs because a correct part is already in contact or a correct load value cannot be acquired (step S43).
4) Load operation start The motor speed and gain are changed, and the descent operation is started (step S44).
A load monitoring task for monitoring the load value of the load cell in real time is started, and monitoring of the load value is started (step S51).
5) Event reception When a load monitoring task detects an error such as when the load cell load value has changed from 0 (step S52) or when a load significantly exceeding the specified load is detected (step S53). Is received (step S45).
6) Acquisition of Z-axis height In order to perform error determination, the Z-axis height at the time of event reception is acquired (step S46).
7) Error determination Error determination is performed using the component type, component outer dimensions, and the like (step S47).
Details of the error determination conditions are shown in Table 1.
8) Pressurization time elapses The above processes 1) to 8) are repeated until the pressurization operation for the specified load time is completed (step S48).
9) Issuance of load monitoring end event When the pressurizing operation for the specified load time is completed, an end event is issued to the load monitoring task, and the load operation is ended (step S49).
10) Error processing Error processing at the time of load error is executed (step S50).

  According to the present invention, in addition to the effect obtained by the first embodiment, since no error occurs in the motor axis coordinate management, an error determination suitable for the application can be made based on the component type, outer dimensions, and the like. Can demonstrate.

<Summary 6>
The present invention detects a load applied from a current value of a Z-axis motor, and controls a Z-axis motor to control the applied load by pressing a component against a board to control the Z-axis stored in advance. By using the cogging torque with respect to the coordinates as a correction value during the pressurizing operation, it is possible to improve the pressurization accuracy when the pressurization is mounted.
Furthermore, the magnitude of the cogging torque is stored according to the Z-axis coordinates.
In addition, by correcting the current value of the Z-axis motor according to the Z-axis coordinates or the rotation angle of the Z-axis motor and reducing the influence of cogging torque, it is possible to improve the pressurization accuracy during pressurization mounting. To do.
In addition, the correction value used as the correction value for cogging torque during the pressurization operation is acquired with the shaft gain and operating speed in the same state as when the pressurization is mounted, so that the accuracy of the correction value is improved and The pressurization accuracy can be improved.
In addition, the cogging torque change due to the difference in operating gain and operating method during pressure mounting is converted by the correction formula using the cogging torque data stored in the storage area, so that the cogging torque due to the change in operating state This makes it possible to maintain the pressure mounting with high accuracy.

(Task)
In the head configuration example in which pressure mounting is performed without using the load detection unit of the first embodiment described above, the load applied to the electronic component can be reduced by directly connecting the Z-axis movable part and the Z-axis motor via a ball screw. It is transmitted directly to the motor.
As a result, the error generating part of the applied load between the component and the motor is limited to the ball screw and the cross roller guide, and by reducing these sliding resistances, the error between the component and the motor is reduced, and in the low load region. It becomes possible to implement the pressure mounting.
Further, the servo motor has a cogging torque that becomes an internal resistance during motor operation, and is generally about 5% to 10% of the rated torque.
Since the Z-axis movable part is designed on the assumption that it operates at high speed, the motor is selected such that the operating torque during normal high-speed operation is 100%.
In the pressure head having the above structure, the operating torque of the motor may be reduced to 10% or less when a low load is applied. As described above, when used in a low output region, there is a problem that correct pressurizing operation cannot be performed due to the influence of cogging torque.

  In the present invention, referring to the torque value of the motor, the influence of the cogging torque of the motor is eliminated in the control for performing the pressure mounting, the pressure mounting operation in the low load region is realized, and the cogging torque is applied in all the load regions. A means for improving the pressurization accuracy is proposed.

(Embodiment 6)
FIG. 26 shows a current value waveform when the Z-axis is lowered at a constant speed under the same conditions as in the pressure mounting in the head configuration described above.
Since the Z-axis is lowered at a constant speed, it can be seen that the current value that is supposed to be a constant value periodically varies in the current value due to the influence of the cogging torque. The waveform in FIG. 26 shows the measurement results for the second time, but the first and second measurement results are stable with no difference.
It is well known that the cogging torque does not change with time.
In the case of the motor used in FIG. 26, the fluctuation range of the current value is -2% to + 1% of the rated torque. When this fluctuation range is converted into a load value, the current value vibrates with a width of about 3N. Will be.
This indicates that an error of about 3N is involved when pressure is loaded.

  In the present invention, these cogging torque values are acquired in advance and the influence of the cogging torque is corrected according to the Z-axis height at the time of pressure mounting, thereby realizing high-precision pressure mounting.

<Cogging torque acquisition process for the target motor>
The cogging torque value of the motor is acquired at the head assembly adjustment stage. The cogging torque value is acquired at a gain setting and operating state that is closer to that during the pressurizing operation.

  The cogging torque value acquisition sequence is as shown in FIG.

1) A load nozzle is mounted (step S61, step S62).
2) Move to the upper limit height of the pressure mounting range and change the shaft gain to the pressure mounting gain (step S63, step S64).
In order to acquire data within the pressure mounting range, the upper limit height of the pressure mounting range is set as the measurement start height (step S65).
3) Lower the Z-axis by an arbitrary movement amount (10 μm), and after completion of the axis movement, wait for the current value to settle, and obtain the current coordinate value and the current value at that time (step S66).
The axis coordinates are not actual coordinates but actual coordinates in consideration of sliding resistance.
4) The acquired current coordinates and current value are stored in the internal table (step S67).
5) The above-mentioned 3) and 4) are repeated until the lower limit height of the pressure mounting range (step S68).
6) The above-described 2) to 5) are performed a plurality of times (5 times) (step S69).
7) The average value of the internal table is adopted as the cogging torque value for each Z-axis coordinate and stored in the storage area (step S70).

  Hereinafter, the data group of the Z-axis coordinates from the upper limit height to the lower limit height and the cogging torque value stored in the storage area is referred to as a cogging torque table.

<Processing during pressure loading operation>
During the pressure mounting operation, the Z axis is lowered at high speed to the upper limit height of the pressure mounting range, and the gain is switched. After that, the pressure is lowered with an operation profile for pressure mounting, and contact with the substrate is detected and a pressure operation with a target load is performed.
The motor current value and the Z-axis coordinate are sampled at a fixed period, the cogging torque value corresponding to the sampled Z-axis coordinate is selected from the cogging torque table stored in the storage area, and the influence of the cogging torque is determined from the acquired current value. The thing excluding is used as the actual applied load.
The cogging torque value to be corrected is calculated by linear interpolation of approximate coordinates from the Z-axis coordinates of the cogging torque table and the actual coordinates when mounted with pressure.
As a result, the influence of cogging torque can be excluded in all the pressurizing regions, so that pressurization accuracy can be improved.

<Correction process according to axis condition>
The pressure mounting operation is intended for pressure mounting from a low load region to a high load region.
Therefore, the gain setting, the operation profile, and the operation sequence differ between the low load region and the high load region.

FIG. 28 shows the result of obtaining the same current waveform by changing the operating gain of the shaft.
Although the period of the waveform does not change, the amplitude of the waveform at the normal gain is larger than that at the load gain.

  Similarly, FIG. 29 shows a change in current value due to a change in the state of the shaft.

“1” indicates a change in the current value when the shaft gain is switched from the setting at the time of high speed operation to the gain at the time of pressure mounting.
Gain switching is performed while the axis is stopped (servo locked).
“2” indicates a change in the current value when the shaft operation is started with the gain at the time of pressure mounting.
There is a change corresponding to the drive current from the current value at the time of stoppage, and it is shown that the influence of cogging appears when the constant speed state is reached.

Thus, the cogging torque value acquired by the gain at the time of operation and the acquisition timing at the time of current changes.
In order to eliminate the influence of these changes, a cogging torque calculation parameter for each shaft state is used.

  Examples of calculation parameters are shown in Table 2.

The calculation parameter includes a cogging torque waveform expansion / contraction coefficient (A) and a shift amount (B), and is calculated in the form of a primary expression Y = AX + B with respect to the cogging torque table value acquired in the adjustment process.
By using this calculation parameter, it is possible to cope with changes in the operation gain and operation sequence without changing the method for obtaining the cogging torque table or storing a plurality of cogging tables for each state.

<Cogging torque correction process using calculation formula>
In the above embodiment, the cogging torque value is acquired for all the pressure mounting ranges, and the approximate value is acquired from the cogging torque table. However, an acquisition method by calculation is also possible.

  As described above, when the cogging torque has a small sliding resistance, the internal structure of the motor comes out as it is as the period of the cogging torque, so the data for one rotation of the motor and the position information (rotation angle of the motor with respect to the Z-axis height). It is possible to calculate an appropriate cogging torque value simply by holding (information), reducing the number of data to be held and shortening the search time for the cogging torque value.

The acquisition method by calculation in the case of a motor in which the cogging torque becomes four cycles when the motor of FIG. 26 makes one rotation is as follows. The cogging torque value of each cycle is represented by table [4] [number of data], and the apex of the cycle serving as the base point is If the acquired Z-axis coordinate is Z1, the cogging cycle is T, and the Z-coordinate from which the cogging torque is acquired is Z0,
Offset value A = (Remainder value of Z1 ÷ (T × 4)) − (T × 4)
Applicable cycle number (table number) = (Z0-A) ÷ T solution Data number = (Z0-A) ÷ the remainder of T divided by the step size of table data, table [table number] [data number ] Is the cogging torque value to be corrected.

  The acquisition flow of the above calculation formula is shown in FIG.

That is, after preparing the table (step S81), the cogging torque table is divided for each period (step S82), and the divided data are stored in the table in order (step S83).
Then, an average value is calculated for each period (table) (step S84), and an application table search formula is calculated (step S85).

Example)
When the cogging torque cycle is 1.5 mm, the data increment is 10 μm, and the value of Z1 is 33 mm, the cogging torque value at 2 mm is A = 33 mm ÷ (1.5 mm × 4) = 5... 3- (1.5 mm × 4)
Therefore A = -3mm
Table number = ((2mm-(-3mm)) ÷ 1.5 = 3 ... 1
So table number = 3
Data number = 1mm ÷ 10μm = 100
Therefore, the torque values stored in the tables [3] and [100] are used as correction values.
Further, when the data step size is large, it is possible to complement the line with the preceding and succeeding data.
Furthermore, although it is inferior in accuracy, it can be converted into a multidimensional expression having a cogging torque value as a Y value and a Z-axis coordinate as an X value and stored.

As described above, according to the present invention, the effects listed below can be obtained.
1) A pressure mounting operation in a low load region can be realized regardless of the magnitude of the cogging torque.
2) The influence of the cogging torque is eliminated, and a highly accurate pressure mounting operation can be realized.
3) The cogging torque value can be dynamically calculated even if there is a change in the generated cogging torque value by switching the operation gain or the operation state.

1 Mounter Device 10 Substrate 13 Mounting Head (Pressure Control Head)
131 Adsorption nozzle 132 Impact buffer spring 18 Electronic component 23 Servo motor

Claims (3)

A mounter device comprising a servo motor that positions the height of a nozzle that sucks a component, and a pressure control head that can control a load pressing the component sucked by the nozzle against a substrate,
The servo motor is also used as a pressurizing source for the nozzle to press the component against the substrate,
The pressurization pressure is variable by the generated output torque of the servo motor generated by the difference between the command level logical coordinate and the actual coordinate of the servo motor ,
The pressure control head of a mounter apparatus, wherein the generated output torque is adjusted by setting a gain corresponding to a set load of the servo motor . A mounter device comprising a servo motor that positions the height of a nozzle that sucks a component, and a pressure control head that can control a load pressing the component sucked by the nozzle against a substrate,
The servo motor is also used as a pressurizing source for the nozzle to press the component against the substrate,
The pressurization pressure is variable by the generated output torque of the servo motor generated by the difference between the command level logical coordinate and the actual coordinate of the servo motor,
The pressure control head of a mounter apparatus, wherein the generated output torque is variable based on a control parameter including a position feedback gain of the servo motor . The pressure control head of the mounter apparatus according to claim 2 , wherein the control parameter is obtained by setting the effectiveness of the integral compensation gain parameter to be low .

Images