JPH09331197A - Component mounting method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce a shutdown of a component mounting device owing to a suction failure when a large component is fed and a manual operation required at a shutdown to an irreducible minimum by a method wherein the replenishment of components to a specific component feeder is detected and stored, it is judged whether a component is held or not when a component is taken out, and a takeout operation is stopped when a component is not held. SOLUTION: For a component mounting device, a first process where the replenishment of components to a specific component feeder out of component feeders is detected and stored by a component replenishment memory 48 and a component replenishment switch 4 respectively, a second process where it is judged by a rotator 35, nozzles 36a to 36d, and a line sensor 40 or a video camera 41 whether a component is held already or not when a component is taken out of the specific component feeder just after the replenishment of components is detected, and a third process where a part takeout operation is stopped by a microcomputer control device 42 at a time when it is judged that a component is not held are provided. By this setup, the part mounting device can be prevented from deteriorating in operating efficiency.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a component mounting method for automatically mounting electronic components on a circuit board by a component mounting device.

[0002]

2. Description of the Related Art First, a component mounting apparatus for automatically mounting electronic components on a circuit board will be described with reference to FIG.

FIG. 1 is a structural example showing the structure of a rotary head system which is one of the general forms of a component mounting apparatus. In the component mounting apparatus shown in FIG. 1, electronic components to be mounted are supplied in a form called taping supply, which is stored in a row on a carrier tape 31 for each type, and the carrier tape 31 is wound on a reel 32. It is mounted on the supply device 33. Parts supply device 33
Holds the carrier tape 31 from the reel 32 at the suction position 34
It has a mechanism that allows the electronic components to be drawn out one by one from the suction position 34.

The rotary head portion H1 has nozzles 36a, 36b for adsorbing electronic components around a rotary body 35 which rotates in a direction indicated by an arrow A in the drawing about a vertical axis.
36c and 36d are attached. The gripping of the component by the nozzle is performed by sucking the electronic component from the suction position 34 by the nozzle that coincides with the position of the nozzle 36b by the rotation of the rotating body 35. At this time, in order to attract a predetermined electronic component to the nozzle, the component supply device 33
A driving device including a servo motor 38 and a servo motor driver 39 is mounted on the component supply table 37 for mounting the component supply table 37 so that the component supply table 37 can be moved in the direction in which the component supply device 33 is arranged in parallel as shown by an arrow Z in the drawing. It is connected.

Circuit board (not shown) after picking up electronic components
The presence or absence of an electronic component may be inspected before the upper mounting, and if there is no component, the mounting may be stopped. In the present configuration example, the presence or absence of the electronic component is detected by measuring the height of the lower end of the electronic component adsorbed at the position of the nozzle 36c by the line sensor 40 in which the detection elements are arranged in the vertical direction. As another method, an image of the electronic component adsorbed at the position of the nozzle 36d is captured by the video camera 41, and the presence or absence of the component is detected depending on whether or not an image of a preset area exists in the screen. There is also.

Further, the operation of these series of devices is unified and managed by the microcomputer controller 42. Now, after the electronic components in the carrier tape 31 are sucked by the nozzles from the suction position 34 of the component supply device 33, the carrier tape 31 has its outlet 4
3 and is cut by the cutter 44. Here, when the nozzle fails to pick up the electronic component, the electronic component remaining in the carrier tape 31 is cut by the cutter 44 together with the carrier tape 31. Therefore, if a large component such as SOP or QFP is mistakenly picked up as an electronic component, the cutter 44 will bite into it, and there is a risk that the cutter 44 will be damaged. Normally, when the nozzle fails to pick up an electronic component, it automatically retries to pick up the part and does not stop the operation. However, when a large part such as SOP or QFP fails to pick up, In order to protect the cutter 44, the operation can be stopped at the time when one suction error occurs. This function is called a large component suction error detection function.

A conventional example of a large component suction error detection function in such a component mounting apparatus will be described below with reference to FIGS. 4 and 5. First, an operation program used for operating the component mounting apparatus will be described with reference to FIG.

The operating program shown in FIG. 5A is NC
It is called a program and designates the mounting order of components, the mounting position, and the number of the component supply device for extracting components. Mounting of components is performed sequentially in the ascending order of block numbers. For each block, the component is taken out from the component supply device with the specified component supply device number, and the component is placed at the position specified by the X and Y coordinates of the circuit board. Implement.

The operation program shown in FIG. 5 (b) is called an array program, and specifies the information of the parts housed in each parts supply device. The column of large parts is 1 when the parts accommodated in each component supply device are large parts, and 0 otherwise.

Next, the operation of the component mounting apparatus will be described with reference to FIG.
1 will be described with reference to the flowchart shown in FIG. First, in step 102, it is checked by the part-out memory 45 of FIG. 1 whether or not the part to be picked up next is already judged to be out-of-part. To do. Next, at step 104, it is checked by the component inspection means whether or not there is a pickup error. Here, the line sensor 40 or the video camera 41 of FIG. 1 is used as the component inspection means.

If it is determined in step 104 that there is no suction error, the component is mounted on the board in step 105, and the mounting operation for the component is completed. On the other hand, if it is determined in step 104 that the pick-up error has occurred, the part is stored in the memory (not shown) in the microcomputer control device 42 as the non-pick-up part in step 106. This stored content is also used when determining the next sucked component in step 102, and also used when checking the presence or absence of the non-sucked component in step 111 later.

Thereafter, in step 107, it is checked whether or not the part is a large part by referring to the arrangement program of FIG. 5 (b). Here, if it is found that the large component is mispicked, the process proceeds to step 119, and the operation of the component mounting apparatus is stopped at that time. After this, if the error is cleared in step 120, step 10
It returns to 2 and the operation is restarted.

[0013] Steps 109 and subsequent steps are processes for a pickup error other than for a large-sized component, and it is checked whether or not the component has been picked up for a predetermined number of times in succession, and if so, in step 110, the component is picked up. Is stored in the parts-out memory 45 of FIG. By this step 110, the suction error processing for one component is completed.

Thereafter, in step 111, it is searched from the NC program of FIG. 5A and the memory of the unsucked parts in the microcomputer control unit 42 whether there is any unsucked parts other than the out of parts. If there is, the process returns to step 102 to perform the suction operation for the part. Also, when the suction of all parts other than the part out has been completed,
In step 112, it is checked whether or not there is a part that is out of stock. If there is a part that is out of stock, in step 115 the parts are replenished, and the process returns to step 102 to restart the operation.

Further, if there is no unmounted component or no component breakage, it means that the mounting of the board is completed. Therefore, in step 113, it is judged whether the operation is completed, and if it is continued, the board is mounted in step 114. After the replacement, the process returns to step 102 to move to the mounting operation of the next board.

[0016]

However, in the conventional component mounting method in the component mounting apparatus as described above, the same electronic component is usually picked up consecutively a predetermined number of times as a criterion for the component mounting apparatus to judge the component shortage. In contrast to the condition that a mistake is made, in the large component suction error detection function, it is assumed that an electronic component has been sucked once.
Since the operation is stopped just by making a mistake in picking up an electronic component once, the operation is stopped every time a pickup error occurs until the electronic component actually runs out and is judged to be out of parts. However, there is a problem that the operating rate of the component mounting apparatus is reduced.

Further, when the supply component is a large component, when a carrier tape feeding defect occurs in the component supply device, the carrier tape retainer of the component supply device is pushed up because the component thickness is large. There is. Therefore, if a suction error occurs due to such a feed error, if the component supply table is automatically moved to the standby position as in the case of a normal component shortage, if the carrier tape retainer is pushed up, There is a risk that the carrier tape retainer may interfere with other parts. For this reason, when a large component is picked up by mistake, it is common to stop the parts supply table on the spot. The table stops on the spot, and it is necessary to manually move the stopped component supply table to the standby position, which also reduces the operating rate of the component mounting apparatus. It also had problems.

The present invention solves the above-mentioned problems. Even when a large component is supplied, the operation provided when the component is sucked is not damaged without damaging the cutter provided for cutting the carrier tape. A component mounting method that can prevent the manual operation for stopping and moving the component supply table to the standby position at that time to be kept to the minimum necessary and prevent a decrease in the operating rate of the component mounting apparatus. provide.

[0019]

In order to solve the above-mentioned problems, the component mounting method of the present invention is designed to collect a plurality of types of components from a plurality of component supply devices, each of which is assembled and stored. When it is determined that the component is taken out and gripped, and then the component is gripped, the component mounting device that automatically mounts the component is set to a specific component supply device of the plurality of component supply devices. The first step of detecting and storing that the part has been replenished, and whether the part is already gripped when the part is taken out from the specific part supply device immediately after the part replenishment detected in the first step A method of executing a second step of determining whether or not a part and a third step of stopping the operation of taking out the part at the time when it is determined that the part is not gripped; To do.

According to this method, even when a large-sized component is supplied, the cutter provided for cutting the carrier tape after the component is adsorbed is not damaged, and the operation is stopped at the time of adsorption failure and the component supply table waits at that time. It is possible to prevent the manual work for moving to the position from occurring every time and to suppress it to a necessary minimum, and to prevent a decrease in the operating rate of the component mounting apparatus.

[0021]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the component mounting method according to claim 1 of the present invention, a plurality of types of components are collected for each type, and the components are taken out from a plurality of component supply devices that are respectively stored and gripped. After that, when it is determined that the component is gripped, the component is automatically replenished to the component mounting device, and the component is replenished to a specific component supply device of the plurality of component supply devices. The first step of detecting and storing the fact that it is stored, and determining whether or not the component is already gripped when taking out the component from the specific component supply device immediately after the component replenishment detected in the first step A method of executing a second step of performing the above step and a third step of stopping the taking-out operation of the component at that time when it is determined that the component is not gripped in the second step.

According to a second aspect of the present invention, there is provided a component mounting method comprising: a component gripping means for picking up a component from a plurality of component supply devices, each of which stores a plurality of components of a plurality of types and stores the components. Mounting control means for controlling the order in which the parts are taken out and the order in which the parts are taken out from the plurality of parts supplying devices; and the parts are replenished to a specific parts supplying device among the plurality of parts. A component replenishment detection unit that detects and stores the component, and a component detection unit that determines whether or not the component gripping unit grips the component after the component is taken out. A first step of detecting that a component has been replenished in the specific component supply device by the replenishment detection means and storing the component, and a specific step immediately after the component replenishment detected in the first step. When taking out a component from the product supply device by the component gripping means, a second step of determining whether or not the component gripping means is gripping the component by the component detecting means, and a component gripping means in the second step When it is determined that the component is not gripped, the mounting control means executes a third step of stopping the component taking-out operation at that time.

In the component mounting method according to the third aspect, as the third step according to the second aspect, when the component gripping means does not grip the component, the mounting control means starts the component from that point onward. A method of stopping the above-described component taking-out operation is performed only for the component supply device which was the target of the taking-out action of the component by the gripping means.

In the component mounting method according to the fourth aspect, as the third step according to the second aspect, when the component holding means does not hold the component, the same type of component as that component is stored. If you have a spare parts feeder, from that point,
The mounting control means suspends the operation of taking out the component from the component supply device, which was the target of the operation of taking out the component by the component holding means, and takes out the component from the spare component supply device.

According to these methods, since large parts such as SOP and QFP have a large adsorbable area due to air pressure, they are unlikely to cause a pick-up error. Therefore, when replenishing parts to the parts feeder, the carrier tape is set. In most cases, it does not work properly and picks up mistakes.These tend to concentrate in the early stages after parts are replenished, and then pick up does not occur until the parts run out. The supply error detection function is made to function only in the initial stage after component replenishment, and if a suction error occurs after that, it is treated as a component shortage, and even for large components, only the parts that are out of component are skipped. Continue to mount the parts.

The component mounting method showing the embodiment of the present invention will be specifically described below with reference to FIGS. 1, 2 and 3. First, an operation program used in the rotary head type component mounting apparatus shown in FIG. 1 will be described with reference to FIG.

The operation program shown in FIG. 3A is NC
It is called a program and designates the mounting order of components, the mounting position, and the number of the component supply device for extracting the components. This NC program is shown in FIG.
It is equivalent to the NC program in (a).

The operation program shown in FIG. 3 (b) is called an array program, and specifies the information of the parts housed in each parts supply device. Of these, the spare number column shows the number of the component supply device to be used as a spare when the component supply device runs out of components.
In the case of (b), there is no spare for the No. 1 component supplying device, No. 3 for the No. 2 component supplying device, No. 3 for the No. 3 component supplying unit, and No. 4 for the No. 4 component supplying unit. It shows that there is no reserve. Further, the column of the component dimension shows the values of the vertical dimension and the lateral dimension of the component. Other items are
This is equivalent to the array program of FIG. 5 (b) shown as a conventional example.

Next, a configuration example of a component mounting apparatus in which the component mounting method of this embodiment is used will be described with reference to FIG. The mechanical structure itself of the component mounting apparatus in which this component mounting method is used is the same as that shown in FIG. 1 described in the conventional example. Therefore, description will be made with reference to FIG. In the component mounting apparatus shown in FIG. 1, the rotating body 35 and the nozzle 36a,
36b, 36c and 36d correspond to the component holding means, the microcomputer control device 42 corresponds to the mounting control means, and the line sensor 40 or the video camera 41 corresponds to the component detection means. The parts corresponding to the parts replenishment detecting means are the parts replenishment memory 48 and the parts replenishment switch 46. The microcomputer control unit 42 writes the number of the component supply device that has run out of components during operation in the component-out memory 45. Further, when the person presses the component replenishment switch 46, the microcomputer control device 42 senses this, and at that time, the component replenishment is performed for all the component supply devices of all the numbers stored in the out-of-component memory 45. The contents of the parts-out memory 45 are copied and copied to the parts supplement memory 48.

Here, the operation program 4 used for operation
When 7 is switched, it is not possible to determine which component has been replenished, so it is considered that all component supply devices used for operation have been replenished, and the numbers of all component supply devices are written in the component replenishment memory 48. To do so. Among the component supply devices determined to be replenished in this way, the large component suction error detection function is performed only for the component supply device that mounts the large component. The determination as to whether or not the component is a large component is made based on whether or not a large component is designated in the array program of FIG. It is also possible to use a method of determining whether or not the component is a large component based on the component dimensions described in the array program.

Next, the operation of the above component mounting apparatus will be described with reference to FIG. 2 showing a flowchart of the component mounting method used in this component mounting apparatus according to the present embodiment. First, when starting the operation of the component mounting device, it is not possible to determine which component is immediately after replenishment,
In step 1, it is assumed that all the parts are parts immediately after replenishment, and the numbers of all the part supply devices are stored in the part replenishment memory 48 of FIG. Step 1 is an example corresponding to the first step of each claim of the present invention.

Next, the mounting operation of each component is started. First, in step 2, it is checked by the component-out memory 45 of FIG. If the product is not out of stock, the component is sucked and gripped by the nozzle in step 3. Next, at step 4, the component inspection means checks whether or not there is a pickup error. Here, when the line sensor 40 of FIG. 1 is used as the component inspection means, if the measured value does not fall within the predetermined range, it is determined that there is a suction error. Further, the video camera 41 shown in FIG.
In the case of using, the suction error is determined when there is no image that matches the component dimension registered in the array program in the imaged screen. In addition to this, as a component inspection means, there is a method of using a normal photosensor, and in this case, a suction error is determined depending on whether the sensor is in an ON state or an OFF state.

If it is determined in step 4 that there is no suction error, the component is mounted on the board in step 5, and the mounting operation for the component is completed. However, in the rotary head type component mounting apparatus as shown in FIG. 1, since the processes for a plurality of components are simultaneously performed in parallel, only the memory that mounting is possible is registered at this point, and when the predetermined timing is reached. The mounting operation may be performed. When the component is mounted, if the component is registered in the component replenishment memory 48, the storage is deleted. As a result, it is determined that it is not immediately after component replenishment at the time of picking up the next and subsequent parts, and the part is excluded from the large component picking error detection function.

On the other hand, if it is determined in step 4 that there is a pick-up error, the part is stored in a memory (not shown) in the microcomputer control unit 42 as a non-pick-up part in step 6. This stored content is used when determining the next suction component in step 2, and also used when checking the presence or absence of the non-suction component in step 11 later. After that, in step 7, whether or not the part is a large part is checked with reference to the array program shown in FIG. Further, when it is determined that the large component is picked up by mistake, it is determined in step 8 whether or not it is immediately after the component replenishment by whether or not it is registered in the component replenishment memory 48 of FIG. Here, the definition immediately after the component replenishment may be the first suction time after the component replenishment or the suction time within a predetermined number of times after the component replenishment.

Of the above processing, steps 4 to 7 correspond to the second step of each claim in the present invention. If it is determined in step 8 that the component has just been replenished, it corresponds to the case where a large component suction error has occurred, and the subsequent abnormal processing is performed.

Here, in the case corresponding to claim 1 and claim 2, step 1 is performed through the route of a in the figure.
Then, the operation of the component mounting apparatus is stopped at that point. The steps 8 and 19 are processes corresponding to the third step of claims 1 and 2. After that, if the error is released in step 20, the large-sized component that has been picked up in mistake in step 16 is considered to be replenished and registered in the component replenishment memory 48 of FIG. 1, and the process returns to step 2 to restart the operation. As a result, even if the component is picked up next time, it is determined that the component has just been replenished, and the operation is stopped at that point. The step 16 is an example corresponding to the first step of each claim of the present invention.

If it is determined in step 8 that the parts have just been replenished and the case corresponds to claim 3, the route of b in the figure is stopped in order to stop the parts taking out operation of only that part. Then, the process proceeds to step 10, and the part is considered to be out of parts and is stored in the out-of-parts memory 45 of FIG. As a result, even if the next attempt is made to pick up the part, the part that has run out of parts is skipped in step 2, so it is possible to prevent the part from being picked up. This step 8 is a process corresponding to the third step of claim 3.

If it is determined in step 8 that the parts have just been replenished, and the case corresponds to claim 4, then, in order to perform the taking-out operation from the spare parts, the route of c in the figure is taken. Then, the process proceeds to step 17. Step 17
Then, it is checked by the arrangement program shown in FIG. 3 (b) whether or not there is a spare component supply device. If there is no spare, the process proceeds to step 19 and the operation is stopped. If there is a spare, in step 18, the component supply device number from which the component is taken out next time is switched to the spare number. The steps 8, 17, and 18 are processes corresponding to the third step of claim 4.

In the case of a component mounting apparatus having a plurality of component supply tables, as a method of switching to a spare component supply apparatus, a method of exchanging the component supply table is possible in addition to the method of referring to the above array program. Is. In this method, the component supply devices are set in the same arrangement in each component supply table, and if a large component adsorption error occurs in a certain component supply table, switching to another component supply table from the time when the next component is taken out. The operation is continued. Since a current component mounting apparatus generally has a plurality of component supply tables, this method can be easily applied.

If it is determined in step 8 that it is not immediately after component replenishment, even if it is a large component, the process is the same as for normal components, and the process proceeds to step 9. In step 9, it is checked whether or not the component has been sucked for a predetermined number of times in succession, and if so, the component is stored as a component out in the component out memory 45 in FIG. 1 in step 10. . By this step 10,
The suction error processing for one component ends.

After that, in step 11, it is determined whether or not there is any non-adsorbed component other than the component exhaustion, as shown in FIG.
The C program and the memory of the non-adsorbed component in the microcomputer control device 42 are searched, and if there is the non-adsorbed component, step 2
Then, the suction operation is performed on the part. Further, when the suction of all the parts other than the parts out has been completed, it is checked in step 12 whether there are any parts out of the parts. If there are parts out, the parts replenishment is waited in step 15, and after the replenishment is completed, the parts are replenished in step 16. The stored component supply device number is stored in the component replenishment memory 48 in FIG. 1, and then the process returns to step 2 to restart the operation. Note that this step 16 is a process corresponding to the first step in each claim of the present invention.

If there is no unmounted component or no component cut, it means that the mounting of the board is completed.
In step 13, it is determined whether or not the operation is completed. If the operation is to be continued, the boards are replaced in step 14, and then the process returns to step 2 to move to the mounting operation of the next board.

As described above, the large component supply error detection function for stopping the operation for each suction error is made to function only in the initial stage after the component replenishment, and if a suction error occurs thereafter, it is treated as a component shortage. Then, even for a large component, it is possible to skip only the component that has run out and continue mounting other components.

As described above, even when a large-sized component is supplied, the cutter provided for cutting the carrier tape after the component is adsorbed is not damaged and the operation is stopped at the time of adsorption failure and the standby position of the component supply table at that time. It is possible to prevent the manual work for moving to the component from occurring every time and to suppress it to the minimum necessary, and to suppress the decrease in the operating rate of the component mounting apparatus.

[0045]

As described above, according to the present invention, the cause of a suction error of a large component can be distinguished between a case where a carrier tape feeding failure occurs due to a component replenishment error and a case where a simple component breakage occurs. , It is possible to stop driving only in the former case of danger.

Therefore, even when the large component suction error detection function is used, unnecessary operation stop can be eliminated. In addition, even if a large component is picked up by mistake, there is no risk of equipment damage unless the component supply device is used. However, the stoppage due to a mistake in picking up a large component can be made to occur only once at the end of mounting the board.

Therefore, unnecessary operation stop can be eliminated more efficiently than the above. In addition, even if a large part is mistakenly picked up, operation will continue as long as there are spare parts, so when the parts including the spare parts are exhausted, the parts will be replenished and checked all at once. be able to.

Therefore, it is possible to eliminate the occurrence of the operation stop itself due to a mistake in picking up a large component, and to eliminate the unnecessary operation stop dramatically and efficiently. Due to the above, even when supplying large parts, the cutter provided for cutting the carrier tape after picking up parts is not damaged and operation is stopped at the time of pick-up error and the part supply table is moved to the standby position at that time. It is possible to eliminate the need for manual work for each time and to keep it to the minimum necessary, and to suppress the decrease in the operating rate of the component mounting apparatus.

[Brief description of drawings]

FIG. 1 is a configuration example of a general component mounting apparatus.

FIG. 2 is a flowchart showing a component mounting method according to the embodiment of the present invention.

FIG. 3 is an operation program of the component mounting apparatus according to the same embodiment.

FIG. 4 is a flowchart showing a conventional component mounting method.

FIG. 5 is an operation program of the component mounting apparatus in the conventional example.

[Explanation of symbols]

 35 Rotating body 40 Line sensor 41 Video camera 42 Microcomputer control device 45 Parts out memory 46 Parts replenishment switch 36a, 36b, 36c, 36d Nozzle

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Kazuo Nagae 1006 Kadoma, Kadoma City, Osaka Prefecture Matsushita Electric Industrial Co., Ltd.

Claims (4)

[Claims] 1. When a plurality of types of components are collected for each type and stored in a plurality of component supply devices, the components are picked up and held, and then, when it is determined that the components are held, A first step of detecting and storing that a specific component supply device among the plurality of component supply devices has been replenished with a component by a component mounting device that automatically mounts the component; Secondly, when the component is taken out from the specific component supply device immediately after the component replenishment detected in the process, it is determined whether or not the component is already gripped.
A component mounting method that executes a process and a third process of stopping the component take-out operation at the time when it is determined that the component is not gripped in the second process. 2. A component holding means for taking out a plurality of types of components from a plurality of component supply devices, each of which collects a plurality of types of components and stores the components, and the operation of performing the component removal operation by the component holding means and the plurality of components. Mounting control means for controlling the order in which the components are taken out from the component supply device, and component replenishment detection for detecting and storing that the component is replenished to a specific component supply device Means and a component detecting device for determining whether or not the component gripping means is gripping the component after the component is taken out, the component replenishment detecting means supplies the specific component to the component mounting apparatus. A first step of detecting and storing that a part has been replenished to the apparatus, and immediately after the part replenishment detected in the first step, the part gripping means from the specific part supply device When taking out a component by means of the second step, the component detecting means determines whether or not the component holding means is holding the component, and in the second step it is determined that the component holding means does not hold the component. If so, the mounting control means executes the third step of stopping the picking operation of the component at that time. 3. As a third step, if the component gripping means is not gripping the component, from then on, the mounting control means is the target of the component pick-up operation by the component gripping means. The component mounting method according to claim 2, wherein the picking-out operation of the component is stopped only for the above. 4. As a third step, if the component gripping means does not grip the component, and if there is a spare component supply device accommodating a component of the same type as that component, mounting is started from that point. The component according to claim 2, wherein the control unit suspends the operation of taking out the component from the component supply device, which is the target of the operation of taking out the component by the component gripping unit, and takes out the component from the spare component supply device. How to implement.