JP2002299369A - Bonding device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent an operator from forgetting a registration position and performing registration which is not for a designated position but for another position, when carrying out registration to a fixed point of an IC chip or the like, which is not detected by a fully automated mode, by operating a manupulator by switching to a semi-automatic mode. SOLUTION: An operator performs registration for an accurate position, while observing a reference image by displaying an image of a fixed point on an IC chip 6, which is stored in advance during registration, as a reference image in a monitor.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a pad (electrode) on an IC chip as a part to be bonded and a lead frame or the like as a part to be bonded to which the IC chip is adhered in a process of assembling a semiconductor device. The present invention relates to a bonding device for connecting external leads formed on a bonding wire with a bonding wire or the like.

[0002]

2. Description of the Related Art Conventionally, a wire bonding apparatus as shown in FIG. 9 has been known as a bonding apparatus for connecting pads (electrodes) on an IC chip to external leads.

As shown in FIG. 9, a wire bonding apparatus includes a camera 1 as an image pickup means having a camera head 1a, an optical lens 1b (hereinafter, referred to as a lens), and a bonding arm 2 having a capillary 2a mounted on a tip thereof.
A bonding head 3 for driving the bonding arm 2 up and down, a positioning means for mounting the bonding means and the image pickup means comprising the bonding arm 2 and the bonding head 3 and two-dimensionally moving and positioning the bonding means in the X and Y directions. And a transfer device 5 having a bonding stage 7 on which a bonding operation is performed by a bonding head 3 including a capillary 2a, a bonding arm 2, and the like.
The bonding stage 7 has a heater 7a for heating a lead frame L / F or the like on which a plurality of IC chips 6 are stuck and arranged in the longitudinal direction.

As shown in FIG. 9, the wire bonding apparatus includes an image processing device 30 for receiving an image signal from the camera 1, a monitor 9 for receiving an output from the image processing device 30, and a control device including a microprocessor. 10 and X
A manipulator 17 for outputting a signal for manually moving the Y table 4 to the control device 10;
0 and the bonding head 3 and X
And a drive device 11 for issuing a drive signal to the Y table 4. The control device 10 includes an operation switch 1
8 are connected.

[0005] As shown in FIGS. 9 and 10, in the wire bonding apparatus, the lead frame L＼F is loaded by the transfer device 5 onto the bonding stage 7 which is heated at the start of the work, and the first IC chip 6 is positioned at the bonding work position. In this state, the bonding operation for the earliest IC chip 6 is performed by operating the bonding arm 2 and the XY table 4. The IC chip 6 has, for example, a square shape, a number of pads 6a arranged side by side along the outer periphery of the upper surface thereof, and is attached to a land portion 14 formed on the lead frame L # F.
Then, these pads 6a are provided on the lead frame L $ F.
Leads 20 corresponding to each are provided.

Next, self-teaching performed in advance of the bonding operation will be described. The self-teaching is for setting various conditions relating to the bonding positions of the pads 6a to be bonded and the leads 20 and the like.

As shown in FIG. 9, the operator operates the manipulator 17 to move the XY table 4 while viewing the image based on the image pickup signal from the camera 1 on the monitor 9, and provides the image on the screen of the monitor 9. When the intersection point O of the cross line 9a is aligned with at least one fixed point on the IC chip 6 and alignment is performed at two points, as shown in FIG. 9 and FIG. O
Are set to two fixed points on the IC chip 6 set by self-teaching, and the positions of the XY table 4 at this time are defined as a first reference point a and a second reference point b. The operator operates the manipulator 17 to move the camera 1 on the XY table 4 so that the intersection O of the cross line 9a provided on the screen of the monitor 9 is fixed on the IC chip and the center of the pad. The operation for adjusting to the center point of the lead and the like is hereinafter referred to as alignment, and in this alignment, X
The case where the eye is adjusted to the first reference point a which is the position of the Y table 4 is referred to as a first eye alignment point, and the case where the eye is adjusted to the second reference point b is referred to as a second eye alignment point. As is clear from FIG.
The position of the optical lens 1 of the camera 1 is different from the position of the coordinates of the XY table. When viewed from the screen of the monitor 9, alignment is performed, and the position coordinates of the XY table at that time are the first reference point a. Or it is the second reference point b.

Then, X at a fixed point on the IC chip 6 is
The position coordinates of the first reference point a and the second reference point b of the Y table 4 are stored in a memory (storage device) in the control device 10,
Also, the first reference point a and the second reference point b of the XY table 4
The image centered on the fixed point of the IC chip 6 in the window 9b on the monitor 9 on the monitor 9 is a reference pattern (the monitor 9 shown in FIG. 9).
It is stored in the memory (storage device) of the image processing device 30 as the upper window 9b).

Next, the above-described alignment at the first reference point a and the second reference point b is alignment of the IC chip 6, and in addition, zero or one or more points on the lead frame L / F. It is necessary to make eye contact. As shown in FIG. 10, when the number of pads is two and the number of leads is one, the position of the XY table 4 which is a fixed point set on the lead frame L / F by self-teaching is set as a third reference point c. Is the third reference point c at the position of the XY table 4 described above.

The position coordinates of the third reference point c of the XY table 4 are stored in a memory in the control device 10, and
An image centered on a fixed point on the lead frame L / F in the window 9b on the monitor 9 at the third reference point c of the XY table 4 is stored in the memory of the image processing device 30 as a reference pattern.

As described above, all the pads 6a and the leads 20 are aligned, and the position coordinates of the bonding points are stored in the memory in the control device 10.

Next, the operation of wire bonding by the bonding apparatus will be described.

In the wire bonding operation, first, I
The shift amount between the C chip 6 and the lead 20 is detected.

As shown in FIG. 10, the lead frame L /
The position of the IC chip 6 on which the plurality of IC chips 6 are arranged and adhered in the longitudinal direction with the conductive bonding material is shifted from a predetermined position to be adhered. There is a predetermined position (6 'indicated by a broken line in FIG. 10) of the IC chip 6, and the above-mentioned first reference point a and second reference point b also exist as fixed points preset by self-teaching. When the lead frame L / F is transferred onto the bonding stage 7 by the transfer device 5, the position of the lead frame L / F on the bonding stage 7 may vary in the X direction.

Therefore, before performing the wire bonding operation, the positions of two fixed points on the IC chip 6 set by self-teaching are detected, and the amount of deviation of the IC chip 6 from a predetermined position is calculated.

The amount of displacement of the IC chip 6 is calculated by moving the camera 1 mounted on the XY table 4 to the first reference point a and the second reference point a.
After the movement to the reference point b, after the movement of the XY table 4 is completed, the camera 1 captures an image of the IC chip 6 and inputs the captured signal to the image processing device 30 for image recognition to perform the first reference point a and the second reference point a. Each shift amount (ΔXa, ΔX in FIG. 10) from the reference point b to a fixed point (a ′, b ′ shown in FIG. 10) on the IC chip 6
Ya and ΔXb, ΔYb). The position of the pad 6a to be bonded of the IC chip 6 is calculated by calculating the bonding position coordinates of each pad 6a stored in advance based on the data of the displacement amount.

The lead 2 serving as a second bonding point
For the position of 0, the position of c ', which is one fixed point on the lead frame L / F, is detected, and the amount of deviation from a predetermined position on the lead frame L / F is calculated. That is, the shift amount of the lead frame L / F can be determined by moving the camera 1 mounted on the XY table 4 to the third reference point c set in advance by self-teaching, and moving the lead frame L / F from the third reference point c.
The shift amount up to c ′, which is a fixed point on / F, is determined. The position of the lead 20 is determined by calculating the bonding position coordinates of the lead 20 stored in advance based on the shift amount.

After detecting the positions of the pads 6a and the leads 20 of the IC chip 6 described above, the XY table 4 is operated to move the capillary 2a into which the wire having the ball formed at the tip of the bonding arm 2 is inserted into the bonding position. Based on the coordinates, it is positioned directly above the pad 6a, and the bonding head 2 shown in FIG.
Is driven to lower the capillary 2a and crush the ball on the pad 6a to perform thermocompression bonding. At this time, the capillary 2a is
Excitation is also performed.

When the connection to the first bonding point is completed, the raising / lowering operation of the bonding arm 2 and the operation of the XY table 4 are performed as appropriate, and the wire is pulled out according to a predetermined loop control. Thus, bonding to the second bonding point on the lead 20 is performed, and the wire is pressed onto the lead 20. Thereafter, the wire is cut in the process of raising the bonding arm 2 to a predetermined height, and the connection between the set of pads 6a and the leads 20 is completed. Thereafter, the above-described series of operations is repeated for each pad 6a provided on the IC chip 6 and each lead 20 provided correspondingly.

[0020]

As described above, in the conventional wire bonding apparatus, when performing wire bonding, the first reference point a and the second reference point b previously set by self teaching. And, after the camera 1 mounted on the XY table 4 is moved to the third reference point c and the movement of the XY table 4 is completed, the camera 1 images the IC chip 6 and the lead frame L / F to be bonded, The image pickup signal is input to the image processing device 30 to perform image processing, and a fixed point (in this case, a ′) on the IC chip 6 from each of the first reference point a, the second reference point b, and the third reference point c. , B '
) And a fixed point on the lead frame L / F (in this case, c ′) is obtained, and the previously stored bonding position coordinates of the pad 6a and the lead 20 are corrected based on the shifted amount. The wire is bonded to the corrected position of each pad and the position of each lead 20.

However, when the fixed points on the IC chip 6 (in this case, a 'and b') are outside the field of view of the camera 1, or when the fixed points on the surface of the IC chip 6 (in this case, a ', b')
The contrast of the pattern near (b ′) changes, and the image of the camera 1 becomes the reference pattern (in this case, the window 9).
If it is different from b), the image processing device 30 may not be able to detect the amount of deviation to a fixed point on the IC chip 6 (in this case, a ′ and b ′). That is, when the contrast of the pattern near the fixed point on the surface of the IC chip 6 is significantly changed, the image processing device 30 cannot recognize the reference pattern because the identity with the reference pattern is lost. Therefore, when the image processing apparatus 30 cannot detect such a shift amount, the wire bonding apparatus stops the operation by issuing an error signal to the outside. In this case, as an error process, the operator operates the manipulator 17 to perform adjustment on a fixed point that cannot be detected.

However, the operator, for example,
There is a case where the alignment position of the pad, which is a fixed point on the C chip 6, is forgotten or the other pad is aligned instead of the designated pad. In particular, when the operator is changed, such a situation is likely to occur. When the operator positions on a pad other than the designated one, an incorrect shift amount is detected, and the position of the pad is not accurately calculated. The position may deviate from the center of the pad, and the ball and the wiring pattern of the IC chip 6 may be short-circuited, which may cause a failure of the IC chip.

Accordingly, the present invention provides a method of displaying an image of a fixed point on an IC chip stored in advance at the time of alignment on a monitor as a reference image so that an operator can perform alignment at an accurate position while viewing the reference image. It is an object of the present invention to provide a bonding apparatus capable of performing the following.

[0024]

SUMMARY OF THE INVENTION A bonding apparatus according to the present invention comprises: a bonding means for performing bonding on a component to be bonded; and a positioning device by two-dimensionally moving the bonding means relative to the component to be bonded. Positioning means for performing, an imaging means for imaging the part to be bonded mounted on the positioning means, an image storage means for receiving an image signal from the imaging means and storing an image of the part to be bonded, and a part to be bonded A bonding device comprising: a shift amount detecting means for detecting a shift amount of the fixed point above; and a display means for displaying a video signal from the image storage means. The imaging signal from the imaging unit at the reference position of the bonding component is cut as image data. Image extracting means, image storing means for storing the image data extracted by the image extracting means as a reference image, and the imaging signal from the imaging means after the positioning means has moved to a reference point position. Image superimposing means for superimposing the reference image stored in the image storage means.

In the bonding apparatus according to the present invention, the image cutout means cuts out a specified range as image data from an image pickup signal including a fixed point of the part to be bonded from the image pickup means.

In the bonding apparatus according to the present invention, the image storage means attaches a registration mark to image data of an image pickup signal from the image pickup means at a reference position of a part to be bonded and stores it as a reference image. It is.

Further, the image superimposing means of the bonding apparatus according to the present invention displays the reference image stored in the image storage means at a designated position of the display means.

In the bonding apparatus according to the present invention, the image pickup signal from the position where the image pickup means of the bonding apparatus is aligned is converted into image data, and the converted image data and the reference of the component to be bonded stored in the image storage means are converted. It has a coincidence detecting means for detecting the coincidence with the reference image at the position.

[0029]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Referring to the drawings, an embodiment of a bonding apparatus according to the present invention will be described below. Parts having the same configuration and function as those of the conventional wire bonding apparatus shown in FIGS. 9 and 10 will be described using the same reference numerals. FIG. 1 is a block diagram showing a configuration of a bonding apparatus according to the present invention, FIG. 2 is a block diagram of a circuit configuration of the image recognition device shown in FIG. 1, FIG. 3 is a block diagram of a configuration of an image superimposing circuit, and FIG. FIG. 7 is a diagram showing a relationship between a control signal for controlling a switch circuit of an image superimposing circuit on a monitor display area of a monitor and an analog signal of a D / A converter displayed on the monitor.

As shown in FIG. 1, the bonding apparatus according to the present invention has a lens 1b attached to the tip and a camera head 1a having a built-in image pickup device, and serves as an image pickup means for picking up an image of a part to be bonded. Camera 1 and
A bonding head 3 as a bonding means including a capillary 2a and the like attached to the tip of the bonding arm 2, and an XY table 4 as a positioning means capable of moving the camera 1 and the bonding head 3 in two-dimensional directions of X and Y directions. And a transport device 5 for transporting a lead frame L / F on which an IC chip 6 as a part to be bonded, which is imaged by the camera 1, to a bonding stage 7, and image recognition for receiving an output from the camera 1 and performing image recognition An image recognition device 8 as a means, a monitor 9 as a display device for receiving an output from the image recognition device 8, a microprocessor for generating a control signal to a driving device 11 as a driving device, and a memory for storing information ( The XY table 4 and the control device 10 including the storage device) are manually moved. A manipulator 17 for outputting a signal for the control to the control device 10 and a drive device 11 as drive means for issuing drive signals to the bonding head 3 and the XY table 4 in response to a command signal from the control device 10 are provided. The control device 10 includes an operation switch 18
Is connected.

As shown in FIG. 2, the image recognition device 8 includes a clock signal generation circuit 8a and a clock signal generation circuit 8a.
A camera control signal generating circuit 8b for outputting a camera control signal such as a horizontal and vertical synchronizing signal to the camera 1 based on a clock from the control circuit 8f and a signal from the control circuit 8f; an amplifier 8c for amplifying an image pickup signal from the camera 1; A converter 8e that samples a signal output from the converter and converts it into a multi-valued digital value (image data), and a range specified from the image data output from the converter 8e (for example, according to the embodiment of the present invention). According to the method, the image data output from the converter 8e is cut out based on the horizontal and vertical cut-out start times and the cut-out end times specified by the control circuit 8f.) Image clipping circuit 8k and converter 8e
An image processing circuit 8d for performing image recognition such as matching of image data output from the image with a reference pattern to detect a shift amount, and a shift between a reference image cut out by the image cutout circuit and a fixed point of the IC chip 6 or the like. Image memory 8 as image storage means for storing reference patterns and the like for amount detection
g, an image memory control circuit 8q for controlling writing and reading of the image memory 8g, and a control circuit 8f containing a microcomputer for controlling the image processing circuit 8d, the camera control signal generation circuit 8b, the image memory 8g, and the like. When,
An image superimposing circuit 8p as image superimposing means for superimposing image data read from the image memory 8g on a video signal from the amplifier 8c; Circuit 8
and an input / output circuit 8h for communicating with the control device 10.

As shown in FIG. 3, the image superimposing circuit 8p
A D / A converter comprising a switch circuit and a D / A converter.
The converter converts the image data read from the image memory 8g into an analog signal and outputs the analog signal to the input terminal (B) of the switch circuit. The video signal from the amplifier 8c is input to one input terminal (A) of the switch circuit.
The switch circuit controls the input terminals (A) and the input terminals (A) based on a control signal input from the image memory control circuit 8q to the input terminal (C).
It is configured to select the input signal of (B) and output the signal to the output terminal (O).

FIG. 4 shows a control signal for controlling the switch circuit of the image superimposing circuit 8p on the monitor display area of the monitor 9 and an analog signal of the D / A converter displayed on the monitor 9. The relationship is shown, as shown in FIG. 4, between the time Ths in the horizontal control signal as the control signal and the time Tv in the vertical control signal as shown in FIG.
By inputting the high-level signal “1” to the input terminal (C) of the switch circuit during Tve, the image memory 8g
The analog signal of the above reference image is output from the switch circuit.

Further, from Ths to T in the horizontal control signal,
time from Tv to Tve in the time to he and vertical control signal
By changing the time until, the reference image read from the image memory 8g can be displayed at an arbitrary position in the monitor display area. The control for displaying the reference image at an arbitrary position on the monitor display area of the monitor 9 is performed by the control circuit 8f of the image recognition device 8 receiving a control signal from the control device 10 by operating the operation switch 18. 8f
A control signal is output to the image memory control circuit 8q, and the control is performed by the image memory control circuit 8q.

Further, as shown in FIG.
The image processing circuit 8d can perform matching between the image data output from the converter 8e and the data of the reference image stored in the image memory 8g. Therefore, the degree of coincidence can be detected by the image processing circuit 8d by the matching processing between the image data of the imaging signal from the camera 1 and the data of the reference image stored in the image memory 8g.

The memory of the control device 10 shown in FIG. 1 incorporates a control program for controlling each operation such as self-teaching and detection of a shift amount described below.

The self-teaching for setting various conditions for detecting the amount of deviation of the fixed point of the IC chip 6 or the like in the bonding apparatus according to the present invention will be described below. FIG. 5 is a flowchart showing a flow of self-teaching for setting a condition for detecting a deviation amount of a fixed point of an IC chip. The detection of the displacement of the fixed point of the IC chip 6 or the like of the bonding apparatus according to the present invention can be performed by a full auto mode in which the displacement is detected automatically by the image recognition device 8 or by the operator operating the manipulator 17. It can be performed in a semi-auto mode in which the detection is performed. Hereinafter, the automatic detection of the shift amount by the image recognition device 8 is referred to as a full auto mode, and the operation of the manipulator 17 to detect the shift amount by an operator is referred to as a semi-auto mode.

First, the operator operates the manipulator 17 to move the XY table 4 while observing the image based on the image pickup signal from the camera 1 on the monitor 9, and moves the XY table 4 to display the image on the screen of the monitor 9. The intersection O of the line 9a is aligned with a first alignment point (a shown in FIG. 8) as a fixed point on the IC chip 6 for the semi-auto mode (step S1 in FIG. 5).

As shown in FIG. 8, when a first reference point for the semi-auto mode at the position of the XY table 4 shown in FIG.
The second point for the semi-auto mode
The reference point b corresponds to a second matching point as a fixed point on the IC chip 6. The first reference point a and the second reference point b for the semi-auto mode are position coordinates on the XY table 4 and are different from the reference pattern shown in the window 9b of the monitor 9 shown in FIG. .

As shown in FIG. 5, the first alignment point (FIG. 8)
After the completion of the alignment shown in a), the operation switch 18 shown in FIG. 1 is pressed to notify the controller 10 of the completion of the alignment.
After confirming that the operation switch 18 has been pressed, the control device 10 stores the position coordinates of the XY table 4 in a memory in the control device 10 as a semiautomatic first reference point (a shown in FIG. 8). Also, the image in the window 9b on the monitor 9 at the semiautomatic first reference point (a in FIG. 8) is used as a reference image of the first matching point (a in FIG. 8).
Is stored in the image memory 8g (step S2). The control circuit 8f shown in FIG. 2 stores the data for the registration mark around the position corresponding to the intersection O of the cross line 9a of the reference image stored in the image memory 8g.
Write to g (step S3).

Next, the XY table 4 is moved by manipulating the manipulator 17 while observing the image based on the image pickup signal from the camera 1 on the monitor 9, and the intersection of the cross line 9 a provided on the screen of the monitor 9 is obtained. O is aligned with a first detection point (D shown in FIG. 8) as a fixed point on the IC chip 6 for the full auto mode (step S4).

The first detection point for the full auto mode is shown in FIG.
As shown in the figure, the window 9b on the screen of the monitor 9 is positioned by moving the XY table 4 within the frame of each side surrounding the outer periphery of the IC chip 6, and the cross line 9 of the window 9b is
The first reference point for the full auto mode on the XY table 4 at the intersection O of a is set as the first detection point D as a fixed point on the IC chip. As shown in FIG. 8, the second detection point for the full auto mode is to position the window 9b on the screen of the monitor 9 by moving the XY table 4 within the frame of each side surrounding the outer periphery of the IC chip 6. The second reference point for the full auto mode on the XY table 4 at the intersection O of the cross line 9a of the window 9b is defined as a second detection point E as a fixed point on the IC chip. That is, the difference between the first reference point and the second reference point for the semi-auto mode is the first reference point for the full auto mode.
The reference point and the second reference point are points defined by the reference pattern of the window 9b shown in FIG. Thus, FIG.
As shown in (1), the frame of the window 9b is set to be within the frame of each side surrounding the outer periphery of the IC chip 6. This is because when a part or the whole of the window 9b is set to protrude from the frame of each side surrounding the outer periphery of the IC chip 6, the IC 9
This is to prevent the detection accuracy from being affected by the influence of the adhesive or the like to which the chip 6 is attached.

Next, as shown in FIG. 5, after completion of the matching of the first detection point (D shown in FIG. 8), the operation switch 18 is pressed to notify the controller 10 of the completion of the matching. Control device 1
0 is XY after confirming that the operation switch 18 has been pressed.
The position coordinates of the table 4 are stored as a first reference point for full auto mode (D shown in FIG. 8) in a memory in the controller 10 and a monitor 9 for the first reference point for full auto mode (D shown in FIG. 8). The image in the upper window 9b is stored in the image memory 8g of the image recognition device 8 as a reference pattern (step S5). As shown in FIG. 8, the first alignment point a for the semi-auto mode and the first detection point D for the full auto mode are set at different positions.
This can be set at the same point (position) on the IC chip 6. Second alignment point b for semi-auto mode
The same applies to the second detection point E for the full auto mode.

Next, the operator operates the manipulator 17 to move the XY table 4 while observing the image based on the image pickup signal from the camera 1 on the monitor 9, and moves the XY table 4 to the cross line provided on the screen of the monitor 9. Intersection O of 9a
Is aligned with the second alignment point (b shown in FIG. 8) as a fixed point on the IC chip 6 for the semi-auto mode (step S6). After the completion of the second alignment point (b shown in FIG. 8), the operation switch 18 is pressed to notify the controller 10 of the completion of the alignment. After confirming that the operation switch 18 has been pressed, the control device 10 stores the position coordinates of the XY table 4 as a semi-automatic second reference point (b shown in FIG. 8) in a memory in the control device 10. The image in the window 9b on the monitor 9 at the second reference point for semi-auto (b in FIG. 8) is used as a reference image of the second matching point (b in FIG. 8) as the image memory 8g of the image recognition device 8. (Step S7). Then, the control circuit 8f shown in FIG.
Writes the data for the registration mark into the image memory 8g centering on the position corresponding to the intersection O of the cross line 9a of the reference image stored in the image memory 8g (step S).
8).

Next, while manipulating the manipulator 17 to move the XY table 4 while observing the image based on the image pickup signal from the camera 1 on the monitor 9, the intersection of the cross line 9 a provided on the screen of the monitor 9 is obtained. O is aligned with a second detection point (E shown in FIG. 8) as a fixed point on the IC chip 6 for the full auto mode (step S9). After the completion of the alignment of the second detection point (E shown in FIG. 8) for the full auto mode, the operation switch 18 is pressed to notify the controller 10 of the completion of the alignment. The control device 10 includes the operation switch 1
After confirming that the button 8 has been pressed, the position coordinates of the XY table 4 are used as the second reference point for full auto (E shown in FIG. 8) in the memory of the controller 10 and the second reference point for full automatic mode. Window 9b on monitor 9 at point (E shown in FIG. 8)
Are stored in the image memory 8g of the image recognition device 8 as reference patterns (step S10).

With the above operation, the self-teaching for setting the data for detecting the fixed point of the IC chip 6 in the semi-auto mode and the full auto mode is completed.

Next, the operator operates the manipulator 17 to move the XY table 4 while viewing the image based on the image pickup signal from the camera 1 on the monitor 9, and moves the XY table 4 to the cross line provided on the screen of the monitor 9. Intersection O of 9a
Is aligned with the third alignment point (c shown in FIG. 8) as a fixed point on the lead for the semi-auto mode. After the completion of the third matching point (c shown in FIG. 8), the operation switch 18 is pressed to notify the controller 10 of the completion of the matching. After confirming that the operation switch 18 has been pressed, the control device 10 stores the position coordinates of the XY table 4 in the memory of the control device 10 as the third reference point for semi-auto (c shown in FIG. 8). Also, the image in the window 9b on the monitor 9 at the semi-automatic third reference point (c shown in FIG. 8) is used as a reference image of the third matching point (c shown in FIG. 8) in the image memory 8g of the image recognition device 8. Is stored. Then, the control circuit writes the data for the alignment mark into the image memory 8g centering on the position corresponding to the intersection O of the cross line 9a of the reference image stored in the image memory 8g.

Next, the XY table 4 is moved by operating the manipulator 17 while observing the image based on the image pickup signal from the camera 1 on the monitor 9 so that the cross line 9 a provided on the screen of the monitor 9 is displayed. The intersection point O is set to the third detection point (F shown in FIG. 8) as a fixed point on the lead for the full auto mode. After the completion of the alignment of the third detection point (F shown in FIG. 8), the operation switch 18 is pressed to notify the controller 10 of the completion of the alignment. The third detection point (F shown in FIG. 8) as a fixed point on the lead for the full auto mode is also detected from the reference pattern of the window 9b as shown in FIG.

After confirming that the operation switch 18 has been pressed, the control device 10 sets the position coordinates of the XY table 4 as the third reference point for full auto mode (F shown in FIG. 8) in the memory in the control device 10. The image in the window 9b on the monitor 9 at the third reference point for full auto mode (F shown in FIG. 8) is stored in the image memory 8g of the image recognition device 8 as a reference pattern.

Next, the XY table 4 is moved by operating the manipulator 17 to the intersection O of the cross line 9a provided on the screen of the monitor 9, and bonding is performed at a predetermined position on the lead 20 or the pad 6a. The position of the XY table 4 is stored in the memory of the control device 10 as the position coordinates of the bonding point when the alignment is completed.

Thereafter, a predetermined number or all of the pads 6a and the leads 20 are aligned, and the position coordinates of the bonding points are stored in a memory in the control device 10.

With the above, the self-teaching for setting the position coordinates and setting the image data for performing the wire bonding operation is completed.

Next, a description will be given, with reference to the flow chart shown in FIG. 6, of the detection of the shift amount by the operator at the time of error stop in the detection of the shift amount in the full auto mode in the wire bonding operation by the bonding apparatus.

As shown in FIG. 6, the amount of displacement of the IC chip 6 is detected by moving the camera 1 mounted on the XY table 4 to the first reference point for full auto mode (D shown in FIG. 8) ( After the movement of the XY table 4 is completed, the camera 1 captures an image of the IC chip 6 and inputs the captured signal to the image recognition device 8 (step S21). The image recognition device 8 converts an imaging signal from the camera 1 into image data,
The image processing circuit 8d (shown in FIG. 2) performs a matching process between the converted image data and the reference pattern of the first reference point (D shown in FIG. 8) for the full auto mode, thereby obtaining the first reference for the full auto mode. The amount of deviation from the point (D shown in FIG. 8) is calculated (step S22). The shift phenomenon in this case is
As shown in FIG. 10, the IC chips 6 ′ to 6
In this case, the shift amount is calculated.

Next, it is determined whether or not the matching rate in the matching processing in step S22 is equal to or more than a predetermined reference value (step S23). If the matching ratio is equal to or greater than the predetermined reference value, it is determined that the detection of the shift amount is normal, and the detected shift amount is stored (step S24).
The camera 1 mounted on the Y table 4 is moved to step S32
To the second reference point for full auto mode (E shown in FIG. 8). If the matching rate is less than the predetermined reference value, it is determined that the deviation amount is abnormal, and an error signal is output to stop the operation (step S25).

Next, the operator releases the error signal output from the bonding apparatus, and operates the operation switch 18 to switch the detection mode from the full auto mode to the semi-auto mode (step S26).

When the bonding apparatus is switched to the semi-auto mode, the camera 1 mounted on the XY table 4 moves from the first reference point for the full auto mode (D shown in FIG. 8) to the first reference point for the semi-auto mode (see FIG. 8). The process moves to a) shown (step S27). After the movement of the XY table 4 is completed, the image recognition device 8 reads the reference image of the first matching point (a shown in FIG. 8) stored in the image memory 8g, and outputs the video signal from the camera 1 by the image superimposing circuit 8p. Is output to the display control circuit 8m, and an image in which the reference image is superimposed is displayed on the monitor 9 (step S28). FIG. 7 is a diagram illustrating an example in which an image on which a reference image is superimposed is displayed on a monitor.
The reference image is displayed on the monitor 9 by operating the operation switch 18.
It can be displayed at any position above.

Next, the operator operates the manipulator 17 while looking at the reference image on the monitor 9 to move the XY table 4 so that the intersection O of the cross line 9a provided on the screen of the monitor 9 is set to the first position. A registration point (a shown in FIG. 8) is registered (step S29). The operator operates the operation switch 18 and presses the registration completion switch. After this switch is pressed, the image processing circuit 8d of the image recognition device 8 stores the image data output from the converter 8e and the image memory 8g. 1st for semi-auto mode
Matching is performed with the image data of the reference image at the reference point (a shown in FIG. 8), and the image data of the imaging signal from the camera 1 and the first semi-auto mode first data stored in the image memory 8g.
The degree of coincidence with the image data of the reference image at the reference point (a shown in FIG. 8) is detected. The detected coincidence is displayed on the monitor 9 (step S30). At this time, the degree of coincidence may be output to the control device 10 to determine whether the degree of coincidence is equal to or greater than the reference value.

The control device 10 is provided with a semi-auto mode first
The amount of movement of the XY table 4 from the reference point (a shown in FIG. 8) is stored as a deviation amount (step S31).

Next, the camera 1 mounted on the XY table 4 moves to the second reference point for semi-auto mode (b shown in FIG. 8). In the case of transition from step S24, the second reference point for full auto mode (E shown in FIG. 8)
(Step S32). After the movement of the XY table 4, the operation switch 18 is used to switch to the full auto mode, and the shift amount detection operation in the full auto mode is restarted. Further, the operator operates the manipulator 17 while viewing the reference image on the monitor 9 in the semi-auto mode to move the XY table 4 to determine the intersection of the cross line 9a provided on the screen of the monitor 9 as the second joint point. (B shown in FIG. 8).

The coordinates of the bonding position of each pad 6a stored in advance are calculated based on the data of the shift amount, and I
The position of the pad 6a of the C chip 6 to be bonded is calculated.

The lead 2 serving as a second bonding point
As for the position of 0, the position of one fixed point on the lead frame L / F is detected, and the amount of deviation from a predetermined position on the lead frame L / F is calculated. The position of the lead 20 is obtained by calculating the previously stored bonding position coordinates of the lead 20 based on the calculated shift amount.

Based on the calculated position coordinates of the pad 6a and the lead 20, the wire bonding operation is performed for each pad 6a provided on the IC chip 6 and each lead 20 provided corresponding thereto.

As described above, by switching to the semi-auto mode when the detection cannot be performed in the full auto mode, the operator can perform the eye alignment while looking at the reference image and adjust the eye to the correct position.

The bonding apparatus according to the present invention
Although the wire bonding apparatus has been described, the present invention can also be applied to the detection of the position of a lead on an IC chip or a tape carrier in a tape bonding apparatus.

[0066]

As described above, according to the bonding apparatus of the present invention, an image in the vicinity of a fixed point to be aligned of a component to be bonded is registered in advance as image information.
When the fixed point of the part to be bonded is aligned using the image information, by displaying the registered image information on the monitor, the operator can perform the accurate alignment with reference to the displayed image, The occurrence of bonding failure can be prevented.

Further, according to the bonding apparatus of the present invention, even if the operator is changed, the alignment can be accurately performed with reference to the displayed image, so that the operation rate of the bonding apparatus is improved and the bonding process is improved. Can increase the number of semiconductor devices produced.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a bonding apparatus according to the present invention.

FIG. 2 is a block diagram showing a circuit configuration of the image recognition device shown in FIG.

FIG. 3 is a block diagram illustrating a configuration of an image superimposing circuit of the image recognition device.

FIG. 4 is a diagram illustrating a relationship between a control signal for controlling a switch circuit of the image superimposing circuit and an analog signal of a D / A converter displayed on a monitor.

FIG. 5 is a diagram showing a flowchart of self-teaching for setting conditions for detecting a shift amount of a fixed point of an IC chip.

FIG. 6 is a diagram illustrating a flowchart of detection of a shift amount by an operator at the time of an error stop in detection of a shift amount in a full auto mode.

FIG. 7 is a diagram illustrating an example in which an image on which a reference image is superimposed is displayed on a monitor.

FIG. 8 is a diagram showing a position of an IC chip attached to a lead frame and showing a fixed point as a reference point.

FIG. 9 is a block diagram illustrating a configuration of a conventional wire bonding apparatus.

FIG. 10 is a diagram showing a position of an IC chip attached to a lead frame and showing a shift amount of a fixed point of the IC chip from a reference point.

[Explanation of symbols]

 L / F Lead frame 1 Camera 1a Camera head 1b Optical lens (lens) 2 Bonding arm 2a Capillary 3 Bonding head 4 XY table 5 Transport device 6 IC chip 6a Pad 7 Bonding stage 7a Heater 8 Image recognition device 8a Clock signal generation circuit 8b Camera control signal generation circuit 8c amplifier 8d image processing circuit 8e converter 8f control circuit 8g image memory 8h input / output circuit 8k image cutout circuit 8m display control circuit 8p image superimposition circuit 8q image memory control circuit 9 monitor 9a cross line 9b window 10 control Device 11 Drive device 14 Land unit 17 Manipulator 18 Operation switch 20 Lead 30 Image processing device

Claims (5)

[Claims] 1. A bonding means for performing bonding on a component to be bonded, a positioning means for performing relative positioning of the bonding means two-dimensionally with respect to the component to be bonded, and positioning, the mounting means being mounted on the positioning means. Imaging means for imaging the part to be bonded, image storage means for storing an image of the part to be bonded in response to an imaging signal from the imaging means, and a deviation amount for detecting a deviation amount of a fixed point on the part to be bonded. A bonding apparatus comprising: a detection unit; and a display unit that displays a video signal from the image storage unit, wherein the image storage unit transmits the image signal from the imaging unit at a reference position of a part to be bonded taught to the positioning unit. Image cutting means for cutting out the image pickup signal as image data, and the image cutting means Image storage means for storing the image data cut out in step 2 as a reference image, and superimposing the reference image stored in the image storage means on the image pickup signal from the image pickup means after the positioning means has moved to the position of the reference point. A bonding apparatus comprising: 2. The bonding apparatus according to claim 1, wherein the image cutout unit cuts out a specified range as image data from an image pickup signal including a fixed point of the part to be bonded from the image pickup unit. 3. The image storage device according to claim 2, wherein the image data of the image pickup signal from the image pickup device at the reference position of the part to be bonded is marked with a registration mark and stored as a reference image. 2. The bonding apparatus according to 1. 4. The bonding apparatus according to claim 1, wherein said image superimposing means displays the reference image stored in said image storage means at a position designated by said display means. 5. An image pickup apparatus according to claim 5, wherein said image pickup means converts an image pickup signal from a position adjusted by said image pickup means into image data, and converts the converted image data and a reference image at a reference position of a part to be bonded stored in said image storage means. 2. The bonding apparatus according to claim 1, further comprising a coincidence detecting means for detecting the coincidence of the two.