阅读说明：本技术 用于校准组件安装设备的方法 (Method for calibrating a component mounting device ) 是由 哈拉尔德·汉德洛斯 弗洛里安·斯皮尔 于尔根·斯图尔纳 于 2019-05-21 设计创作，主要内容包括：本发明涉及一种用于校准组件安装设备的方法。所述组件安装设备被配置为将组件安装在基板上,所述基板的安装地点不包含局部标记。基板包含附接在其边缘上的全局基板标记或能够用于安装组件的其他全局特征。校准利用校准板(1)被执行,校准板(1)具有在校准板(1)上方二维分布并且被设置有第一光学标记(3)的若干个校准位置(2)、具有第二光学标记(6)的测试芯片(5)、和附接到在用于临时容纳校准板(1)的接合操作台(17)的保持器(9)。除了可能的例外之外,校准板(1)的校准位置(2)的数目和布置以及基板的安装地点的数目和布置彼此不同。(the present invention relates to a method for calibrating a component mounting device. The component mounting apparatus is configured to mount a component on a substrate whose mounting place does not include a local mark. The substrate contains global substrate markings or other global features that can be used to mount components attached on its edge. The calibration is performed with a calibration plate (1), the calibration plate (1) having a number of calibration positions (2) distributed two-dimensionally over the calibration plate (1) and provided with first optical markers (3), a test chip (5) having second optical markers (6), and a holder (9) attached to a bonding station (17) for temporarily accommodating the calibration plate (1). The number and arrangement of the calibration positions (2) of the calibration plate (1) and the number and arrangement of the mounting locations of the substrates differ from each other, with possible exceptions.)

1. A method for calibrating a component mounting device comprising a bonding station (17) and at least one bonding head (14), a first camera (15) and a second camera (16) for placing a component on a mounting location of a substrate, wherein mounting the component on the mounting location of the substrate comprises the steps of:

transferring the substrate to the bonding station (17) and fixing the substrate;

determining global substrate position data characterizing the position and orientation of the substrate; and

Mounting one of the components after the other on a mounting site of the substrate by means of the following steps:

picking up the assembly from a supply unit with the at least one bonding head (14);

Taking an image of the component held by the bond head (14) with the first camera (15) and determining a deviation of the actual position of the component from its target position,

calculating the actual location of the installation site based on the global substrate location data;

calculating a correction vector to be used for the installation site based on the selected calibration data;

-calculating the position to be approached by the engagement head (14); and

Moving the bond head (14) to the calculated position and placing the assembly on the substrate;

And wherein the calibration comprises determining calibration data using the steps of:

positioning a calibration plate (1) in a holder (9) of the joining station (17) and/or fixing a calibration plate (1) to a holder (9) of the joining station (17), the calibration plate (1) having a plurality of calibration positions (2) distributed two-dimensionally over the calibration plate (1) and being provided with first optical markings (3);

-performing the following steps for a number of said calibration positions (2):

picking up a test chip (5) provided with a second optical mark (6) by means of the bond head (14),

Capturing an image of the test chip (5) held by the bonding head (14) with the first camera (15) and determining a deviation of the actual position of the test chip (5) from its target position,

calculating the position to be approached by the bond head (14) for placing the test chip (5) on the calibration position (2),

Moving the bonding head (14) to the calculated position and placing the test chip (5) on the calibration plate (1),

-taking an image of the test chip (5) placed on the calibration plate (1) with the second camera (16),

Determining a difference vector describing a deviation of the actual position of the test chip (5) from its target position;

-assigning correction data based on at least one difference vector to the calibration position (2), wherein the calibration data comprises the calibration position used and the correction data assigned to the calibration position.

2. the method of claim 1, wherein each of the mounting locations of the substrate does not include any markings.

3. Method according to claim 1, wherein the number and arrangement of the calibration positions used in the calibration and the number and arrangement of the mounting locations of the substrate are different from each other, and wherein the calculation of the correction vectors to be used for the mounting locations is performed by means of an interpolation method which calculates the correction vectors to be used on the basis of selected calibration data comprising one or more calibration positions surrounding a current mounting location and the correction data respectively associated with the one or more calibration positions.

4. Method according to claim 2, wherein the number and arrangement of the calibration positions used in the calibration and the number and arrangement of the mounting locations of the substrate are different from each other, and wherein the calculation of the correction vectors to be used for the mounting locations is performed by means of an interpolation method which calculates the correction vectors to be used on the basis of selected calibration data comprising one or more calibration positions surrounding a current mounting location and the correction data respectively associated with the one or more calibration positions.

5. the method of any of claims 1-4, wherein the substrate remains stationary during installation of the assembly and the calibration data covers an area as large as or larger than the substrate.

Technical Field

The present invention relates to a method for calibrating a component mounting device, also referred to as a device for mounting a component.

Background

in the semiconductor industry, such component mounting equipment is referred to as a die bonder or pick and place machine. A "die" is a component. These components are in particular electrical, optical, micromechanical, micro-optical, or electro-optical components, etc., or also semiconductor chips or flip chips.

Disclosure of Invention

The object of the invention is to develop a component mounting device with which a large number of components can be accurately positioned on a large-area substrate which has substrate markings or other global features on its edges only which can be used for positioning. Such large area substrates are, for example, wafers having a diameter of 12 inches or more. The other substrate is a printed circuit board, a ceramic substrate, a metal sheet carrier, a panel, or the like. Such panels have for example dimensions of 0.6m x 0.7m or more.

according to the invention, the above-mentioned object is solved by a method having the features of claim 1, with which a mechanically highly stable component mounting device can be calibrated.

the present invention therefore relates to a method for calibrating a device for mounting a component (also referred to herein as a component mounting device). The component mounting apparatus includes a bonding station and at least one bonding head for placing a component on a mounting location of a substrate, a first camera, and a second camera. The first camera is used to take an image of the component or the test chip picked up by the bonding head and to determine a deviation of the actual position of the component or the test chip from its target position, respectively. The second camera is used on the one hand to determine global substrate position data characterizing the position and orientation of the substrate during the mounting operation and on the other hand to determine deviations of the actual position of the test chip mounted on the calibration board from a target position during the calibration.

the component mounting apparatus further includes a transfer device that transfers one substrate after the other substrate to a bonding station where it is assembled with the component. The bonding station includes a pick and place system that moves the bonding head to the substrate assembly station. To achieve high placement accuracy, the component mounting apparatus is advantageously configured to convey the substrate to the bonding station and then hold the substrate in place throughout the mounting process. In this case, the working area of the bonding head or heads is as large as the substrate or larger.

Since each mounting site of the substrate does not contain a mark defining its position, the position of each mounting site where the bonding head must approach to mount the component is calculated from global substrate position data characterizing the position and orientation of the substrate. Once the substrate has been transferred to the bonding station and secured there, the global substrate position data is determined by the substrate marks provided on the perimeter of the substrate, or if the substrate is free of substrate marks, by specific global features of the substrate such as "flats" and/or "notches" on the wafer.

to ensure that the position approached by the bond head matches the position calculated for the installation site, the component mounting apparatus is calibrated. During the calibration, calibration data is determined. The calibration is performed with a holder, a calibration plate having a number of calibration positions and with one or more test chips. The calibration positions of the calibration plate are two-dimensionally distributed over the calibration plate; they are arranged, for example, in rows and columns. The calibration positions of the calibration plate and the test chip comprise a first optical mark and a second optical mark, respectively, which match each other.

the holder for holding the calibration plate is permanently or removably mounted in the joining console of the component mounting apparatus. The component mounting apparatus and the holder are configured such that the calibration data completely covers the working area of the bond head or heads. A single calibration plate can be used to cover the entire working area or the holder can be designed to receive the calibration plate in sequence in different positions to cover the entire working area. The working area, and thus the area covered by the calibration data, is advantageously as large or larger than the entire substrate, so that the substrate can remain stationary at the same position throughout the mounting process.

during calibration, a single test chip or several test chips are placed at several selected calibration positions, and the deviation of the actual position of each placed test chip from its target position is determined using two cameras and also provided image processing hardware and software. The deviation is preferably determined immediately after the placement of the respective test chip. The deviation can be represented by a difference vector v having at least two components (e.g., having two components v ═ x, y or having three components v ═ x, y, θ), where component x indicates displacement in a first direction, component y indicates displacement in a second direction, and component θ is an angle indicating rotation about the center. A difference vector v with two components is usually sufficient, i.e. then when the angular deviation theta is so small that no disturbing position errors arise.

for example, if the detected difference vectors indicate a non-linear behavior of the axis of movement of the bond head or heads, respectively, the number of selected calibration positions can be increased adaptively during the method.

the holder is preferably configured to hold the calibration plate by vacuum, and the holder and the calibration plate are preferably configured to hold the test chip by vacuum. The holder and calibration plate may also be designed to magnetically hold the calibration plate. The holder and/or the calibration plate and the test chip may also be designed to magnetically secure the test chip.

in another embodiment, the holder consists of only position pins, which are preferably permanently attached to the joining station. In this case, it is not necessary to insert and remove the holder for calibration again. Such a holder is not able to hold the calibration plate, but only to position it.

The calibration plate is preferably a glass plate and the test chip is preferably a glass chip. When such a structure can be produced with extremely high accuracy, the first optical mark and the second optical mark are preferably structures made of chromium. These structures are optically opaque.

the determination of the calibration data comprises, for example, the following steps:

A) Positioning the calibration plate in and/or securing the calibration plate to the holder of the joining station;

B) Performing the following steps C to I for a plurality of calibration positions set on the calibration plate:

C) the test chip is picked up with the bond head,

D) taking an image of the test chip held by the bond head with a first camera and determining a deviation of the actual position of the test chip from its target position,

E) Calculating a position to be approached by the bond head for placing a test chip at the calibration position,

F) Moving the bonding head to the calculated position, and placing the test chip on the calibration plate,

G) An image of the test chip placed on the calibration plate is taken with a second camera,

H) A difference vector v is determined, which describes the deviation of the actual position of the test chip from its target position.

If no deviation is detected, the difference vector v is a zero vector. At a suitable time after step G, the test chip or chips are each removed. Of course, the same test chip is always used. The calibration can be refined by repeating all or some of the steps selected from steps C to H once or several times in order to obtain additional difference vectors.

after all or some of the steps C to H have been performed once or several times, one or more difference vectors exist for each calibration position. Therefore, the following steps still need to be taken:

I) Correction data are assigned to the calibration positions based on the at least one difference vector v.

The following applies to step I for each calibration position used: if steps C to H have been performed only once, the correction data of the calibration position contain a difference vector v. If some of the steps C to H have additionally been performed one or more times, several difference vectors v are available. The correction data may then contain, for example, all or some of the difference vectors v selected according to certain criteria, or the correction data may alternatively contain correction vectors calculated from all or a few of the selected difference vectors v.

if the component mounting apparatus has more than one bonding head, the above-described process is performed for the work area of each of the bonding heads.

if the size of the calibration plate is too small to cover the entire working area of the bonding head or heads, the calibration plate is attached to the holder in various positions, and the above-described calibration process is performed for each position. The different positions are designed to cover the entire working area of the bond head.

After the calibration has been performed, there is calibration data comprising the calibration positions used and the correction data assigned to these calibration positions. The calibration position is defined by a vector w having at least two components (e.g., having two components w ═ w1, w2 or having three components w ═ w1, w2,), where component w1 indicates a position in a first direction, component w2 indicates a position in a second direction, and the component is an angle indicating a rotation around the center. Thus, the calibration data includes the vector w and the correction data for each of the calibration positions used.

The number and arrangement of the calibration positions of the calibration plate 1 and the number and arrangement of the mounting locations of the substrates differ from each other with possible exceptions.

in the case of a component mounting apparatus in which a substrate is transferred from a transfer device to a bonding station and fixed there, for example, by suction vacuum, then assembled with a component, and then transferred away from the bonding station, calibration data covers an area as large as or larger than the substrate.

once the calibration has been completed, the component can be mounted on the mounting site of the substrate using the following steps:

Transferring the substrate to a bonding stage and fixing the substrate;

Determining global substrate position data representing the position and direction of the substrate; and

mounting a component behind other components at a mounting location behind the other components of the substrate by:

Picking up the component from the supply unit using at least one bonding head;

Capturing an image of the component held by the bond head using a first camera and determining a deviation of an actual position of the component from its target position;

Calculating an actual position of the installation site based on the global substrate position data;

Calculating a correction vector to be used for the installation site based on the selected calibration data;

Calculating a position to be approached by the bond head; and

the bonding head is moved to the calculated position and the assembly is placed on the substrate.

Drawings

Detailed Description

The joining station of the component mounting device is provided with a fixed holder or a temporary reception of a holder configured for receiving (receiving) and holding the calibration plate. The calibration plate is a highly stable carrier with extremely precisely defined calibration positions which are arranged two-dimensionally (in particular in rows and columns) over the entire calibration plate. Since glass is transparent and has excellent mechanical and optical properties for this application, the calibration plate and the test chip are preferably made of glass.

the holder is advantageously configured to temporarily hold the calibration plate and the test chip in place. In a preferred embodiment, each calibration position of the calibration plate is provided with a hole, which can be supplied with a vacuum for fixing the test chip with the vacuum. The vacuum is supplied by a vacuum source. In an alternative embodiment, the holder and/or the calibration plate and the test chip are provided with a magnet and optionally a ferromagnetic element, such that the magnetic force affixes the test chip to the calibration plate.

Fig. 1 shows in top view a calibration plate 1, here a glass plate, suitable for use in a calibration assembly mounting device. The illustrated calibration plate 1 contains a large number of calibration positions 2 arranged in rows and columns. Fig. 2 shows such a calibration position 2 of the calibration plate 1 in an enlarged view. Each calibration position 2 contains a first optical marker 3. In this embodiment, each calibration position 2 also contains a hole 4 through the calibration plate 1 to hold a test chip with a vacuum. The hole 4 is preferably located in the center of the corresponding calibration position 2.

fig. 3 shows a test chip 5. The test chip 5 is optically transparent and contains various second optical markings 6. The test chip 5 is preferably composed of glass.

the first optical markers 3 of the calibration plate 1 and the second optical markers 6 of the test chip 5 are preferably of chrome construction and are therefore optically opaque.

For example, the first optical mark 3 on the calibration plate 1 comprises five rings 7 and the second optical mark 6 on the test chip 5 comprises five rings 8, the diameter of the rings 7 on the calibration plate 1 being different from the diameter of the rings 8 on the test chip 5. The mutual distance of the centers of the rings for the calibration plate 1 and the mutual distance of the centers of the rings for the test chip 5 are the same, so that if the test chip 5 is placed correctly on the calibration position 2, the ring 7 of the calibration plate 1 extends concentrically and separately from the ring 8 of the test chip 5 and can therefore be distinguished optically. The first optical marking 3 and the second optical marking 6 may additionally comprise a scale or a vernier (nonius) scale as shown.

Fig. 4 and 5 show in top view and in cross-section an embodiment of a holder 9 designed according to the invention, which firstly holds the calibration plate 1 with a vacuum and secondly applies a vacuum to the holes 4 of the calibration plate 1. The holder 9 has a flat surface 10, on which flat surface 10 the calibration plate 1 can be placed. The flat surface 10 can have a protruding circumferential edge. The flat surface 10 is provided with a first shaft bore 12, which first shaft bore 12 opens into a first chamber 11, which first chamber 11 is arranged below the flat surface 10 and can be supplied with vacuum. The first shaft holes 12 are arranged in such a way that they are aligned with the holes 4 of the calibration plate 1, so that the test chips placed on the calibration plate 1 are held with vacuum. The flat surface 10 also contains a second axial hole 13, which second axial hole 13 opens into a second cavity or groove of the holder 9, which second cavity or groove can be supplied with vacuum in order to fix the calibration plate 1 to the holder 9 also with vacuum. The first chamber 11 can also be divided into several individual chambers, which can be individually supplied with vacuum. With such a subdivision, the vacuum consumption can be reduced if necessary.

In another embodiment, the holder consists only of position pins, which are preferably permanently attached to the joining station. The insertion of a holder for calibration and removal is not necessary in this case. Alternatively, the position pin may be temporarily inserted into the joining station for calibration and removed again after calibration. Such a holder cannot hold the calibration plate but only can position it, and since it is composed of only the position pins, it also does not have a flat surface. In this example, neither the holder nor the test chip is held with vacuum.

fig. 6 schematically shows a part of the component mounting apparatus necessary for understanding the present invention. The bonding station 17 of the component mounting apparatus comprises a pick and place system with at least one bonding head 14, which at least one bonding head 14 places the components on the substrate. Various substrates are used as the substrate. These substrates are conveyed to the bonding station 17 by the conveying means and away from the bonding station 17. The holder 9 according to the invention is arranged in the joining station 17. The holder 9 is preferably fastened in an exchangeable manner, as it is usually only a calibration requirement for the component mounting device. The component mounting device includes a first camera 15 and a second camera 16, and image processing hardware and software. The first camera 15 is used to determine deviations of the actual position of the component picked up by the bonding head 14 from its target position during mounting or deviations of the actual position of the test chip 5 picked up by the bonding head 14 from its target position during calibration. The second camera 16 is used on the one hand to determine global substrate position data characterizing the position and orientation of the substrate during the mounting operation and on the other hand to determine deviations of the actual position of the test chip 5 from its target position during calibration. There are component mounting apparatuses in which the second camera 16 is mounted at a fixed position or movably mounted at the bonding stage, and there are component mounting apparatuses in which the second camera 16 is attached to the bonding head 14. The first camera 15 is located below the travel path of the bonding head 14, and sees the component or test chip 5 from below. The second camera 16 is positioned above the holder 9 so that the substrate or calibration plate 1, respectively, is in its field of view. The holder 9 is preferably black so that it appears as a black background only in the image of the second camera 16 and therefore does not affect the image processing.

fig. 6 shows the holder 9, the calibration plate 1 and the test chip 5 used during calibration. In the normal mounting mode, the substrate is at the position of the calibration board 1 and the component is at the position of the test chip 5.

The method for calibration of the component mounting apparatus according to the present invention uses the above-described apparatus, the calibration board 1 of the component mounting apparatus, the holder 9 for the calibration board 1, and the bonding head 14, the cameras 15,16, and the image processing hardware and software, and the method includes the steps of:

A) Positioning the calibration plate 1 in the holder 9 of the joining station 17 and/or fixing the calibration plate 1 to the holder 9 of the joining station 17;

B) The following steps C to I are performed for several calibration positions 2 provided on the calibration plate 1:

C) The test chip 5 is picked up with the bonding head 14,

D) an image of the test chip 5 held by the bonding head 14 is taken with the first camera 15, and a deviation of the actual position of the test chip 5 from its target position is determined,

E) The position to be approached by the bonding head 14 is calculated, for the precise placement of the test chip 5 at the calibration position 2 of the calibration plate 1,

F) the bonding head 14 is moved to the calculated position, and the test chip 5 is placed on the calibration plate 1,

G) An image of the test chip 5 placed on the calibration plate 1 is taken with the second camera 16,

H) A correction vector v is determined which describes the deviation of the actual position of the test chip 5 from its target position.

after all or some of the steps C to H have been performed once or several times, one or more difference vectors exist for each calibration position. Therefore, the following steps still need to be taken:

I) correction data are assigned to the calibration positions based on the at least one difference vector v.

The correction data for each calibration position comprises all or some of the difference vectors v, for example selected according to some criterion, or the correction data may alternatively comprise correction vectors calculated from all or several of the selected difference vectors v.

The position to be approached by the bonding head 14 in step E is calculated based on: the actual position of the selected calibration position and the deviation of the actual position of the test chip 5 from the target position determined in step D. The actual position of the selected calibration position 2 can be determined, for example, from global markers arranged in the edge region of the calibration plate 1 or from local markers arranged in the region of the selected calibration position 2. In the second case, the following steps can be performed between step D and step E: an image of the selected calibration position 2 is taken with the second camera 16 and the actual position of the selected calibration position 2 is determined.

fig. 7 shows an image taken by the camera 15 of the test chip 5 placed at the calibration position 2 of the calibration plate 1. The image shows both the first optical marking 3 of the calibration plate 1 and the second optical marking 6 of the test chip 5. The image processing hardware and software of the component mounting device are configured to determine the actual position of the optical markers 6 of the test chip 5 relative to the actual position of the optical markers 3 of the calibration plate 1 and to determine deviations of the actual position of the test chip 5 from its target position.

the calibration plate 1 is fixed on the bonding stage by inserting the mentioned holder 9 at a desired position and placing the calibration plate 1 on the holder 9 and fixing the calibration plate 1 with vacuum or magnetic force.

the determination of the actual position or the calibration position 2, respectively, of the test chip 5 or the determination of the deviation of the actual position of the test chip 5 from its target position, respectively, is performed with the aid of image processing hardware and software of the component mounting device.

because the test chip 5 is transparent, the optical markers 6 of the test chip 5 and the optical markers 3 of the underlying calibration positions 2 are visible in the image captured by the second camera 16.

If the holder 9 is provided with a single first chamber 11, the holes 12 of all calibration positions 2 will of course be supplied with vacuum at the beginning of the test procedure. If the holder 9 is provided with several chambers, a vacuum is applied to one chamber after the other chambers and the test chip 5 is placed in a calibration position assigned to the chamber under vacuum. A single test chip 5 can be used. In this case, the bonding head 14 places the test chip 5 on each of the selected calibration positions 2 of the calibration plate 1 in turn according to the above-described procedure. The number of calibration positions 2 used can comprise all calibration positions 2 of the calibration plate 1 or only some selected calibration positions 2 of the calibration plate 1. The test chip or test chips 5 can also be placed several times on selected calibration positions 2 or on all calibration positions 2.

In the normal operating mode of the component mounting device, it is now possible to place components on the substrate site of the substrate with high positional accuracy, i.e. with the following steps:

A2) Transferring the substrate to the bonding stage 17 and fixing the substrate at the bonding stage 17;

B2) Determining global substrate position data; and is

C2) Through steps D2 to H2, after the other components, one component after the other components is mounted on a mounting site on the substrate:

D2) picking up the component from the supply unit with one of the bond head 14 or the bond head;

E2) taking an image of the component held by the bonding head 14 with the first camera 15 and determining a deviation of the actual position of the component from its target position;

F2) Calculating an actual position of the installation site based on the global substrate position data;

G2) Calculating a position to be approached by the bond head; and

H2) the bond head 14 is moved to the calculated position and the component is placed on the substrate.

the global substrate position data characterizes the position and orientation of the substrate and, hence, the position and orientation of the mounting location. The global substrate position data in step B2 is determined from substrate marks provided on the edge of the substrate or, if the substrate does not have substrate marks, from specific global features of the substrate. For example, if the substrate is a wafer, global features such as flats and/or notches of the wafer may be used. To determine global substrate position data, the second camera 16 captures one or more images of the substrate marks or specific features of the substrate, and image processing hardware and software determines the position of the substrate relative to the bonding head 14 or machine coordinates of the plurality of bonding heads 14.

the calculation of the actual position of the installation site in step F2, which also includes its orientation, is based on global substrate position data.

In step G2, the position to be approached by the bond head is calculated based on the deviation of the actual position of the bond head pickup assembly determined in step E2 from its target position, the actual position of the mounting location calculated in step F2, and the correction vector is determined based on the selected calibration data. The number and arrangement of the alignment positions of the alignment plate 1 and the number and arrangement of the bonding stations of the substrates are different from each other with possible exceptions. Therefore, the calculation of the correction vector to be used for the installation site is advantageously performed by an interpolation method that calculates the correction vector to be used based on the selected calibration data including one or more calibration positions around the current installation low point and correction data assigned to the one or more calibration positions.

In the component mounting apparatus in which the substrate is transferred to the bonding station 17 by the transfer means and fixed thereto, and then assembled with these components and transferred away from the bonding station 17, the calibration data covers an area as large as or larger than the substrate.

since placing the test chip 5 on the calibration site 2 can be performed during a long period of time (e.g., throughout the night) without requiring manual work such as cleaning of the calibration plate 1, the method according to the present invention can also be used to test the positional accuracy or effectiveness of the calibration of the component mounting apparatus under long-term influences such as temperature changes, humidity changes, and the like.

Although the method according to the invention has been developed for large area substrates, the method according to the invention can also be used for the assembly of substrates of any size. The method can also be used under conditions where the substrate comprises local marks.

While embodiments and applications of this invention have been shown and described, it would be apparent to those skilled in the art having the benefit of this disclosure that many more modifications than mentioned above are possible without departing from the inventive concepts herein. The invention, therefore, is not to be restricted except in light of the spirit of the appended claims and their equivalents.