JPH04124841A - Bonding device - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed description of the invention] The present invention will be described in the following order.

A. Industrial field of application B6 Overview of the invention C1 Prior art [Figures 8 and 9] D1 Problems to be solved by the invention E9 Means for solving the problems 11 Effects G. Examples [First Figures to Figures 7] a. First embodiment (collet mounting mechanism) [Figures 1 to 4] b. Second embodiment (feeder width changing mechanism) [Figures 5 to 7] H8 Effects of the Invention (A. Field of Industrial Application) The present invention relates to a bonding device, and particularly to a bonding device for bonding a semiconductor pellet to a predetermined position of a lead frame.

(B. Summary of the Invention) In order to automatically position collets and lead frame feeders accurately and in a short time in the above-mentioned bonding apparatus, the present invention recognizes the positions with a video camera and uses the recognition results. The positioning is performed by moving the conveying member by the amount of deviation from the target position using a conveying moving mechanism.

(C, Prior Art) [Figures 8 and 9] A collet that directly sucks and conveys a semiconductor pellet, which is essential for pellet bonding, conventionally has a vertical hole at the tip of a bonding arm a, as shown in Figure 8. It was attached by pushing it in from the bottom and fixing it with set screw C. surface, d is a collet, and e is a semiconductor pellet.

In addition, the lead frame feeder that supplies the lead frame to which the semiconductor pellet e is bonded is now capable of responding to changes in the width of the lead frame by changing the spacing between the rails using a pulse motor, as shown in Figure 9, for example. ing.

In the same figure, f and f are a pair of rails constituting lead frame feeder g, which are arranged parallel to each other with a variable interval between them, and both have threaded holes h in a direction perpendicular to rail f. , the screw hole h has one rotation axis i
Screws j and j formed in are screwed together. One of the threaded holes and the screw j are reverse threaded, so when the rotation axis i is rotated in one direction, the rails f and f
When the rails f and f move toward each other, and rotate in the other direction, the rails f and f move away from each other. n and I2 are gears that transmit rotational force to the motor, and m is a lead frame.

In some cases, the rails f and f were moved manually without using a motor and fixed with screws to accommodate variations in the width of the lead frame.

(D. Problem to be solved by the invention) By the way, the 8th problem
The mechanism shown in Figs. and 9 had the following problems.

First, referring to the collet mounting mechanism shown in Fig. 8, the diameter of the collet d is, for example, 3 mm and 0.01 mm.
Dimensional tolerances between 3mm and 0.01mm smaller values are allowed, and for example, 3mm and 0.01mm smaller for the mounting holes.
Dimensional tolerances between larger values of m are allowed. Therefore, when collet d is installed, the dimensional error is ±0°0.
Since it may be about 1 mm, errors occur in the degree of contact between the collet and the pellet and the position in the X, Y directions, and rotational direction (θ direction). Therefore, each time the collet is replaced, an adjustment is required to correct the error by a skilled worker, and the time required for the adjustment is too long to ignore.

Furthermore, regarding the lead frame feeder width changing mechanism shown in Fig. 9, position reproducibility is poor due to backlash and backlash, and the width needs to be adjusted every time the lead frame width is changed. It took skill and time to do it.

These are problems that cannot be overlooked because they naturally become a factor that hinders the improvement of productivity and, in turn, cause an increase in the cost of semiconductor devices such as ICs and LSIs.

The present invention has been made to solve these problems, and an object of the present invention is to enable automatic positioning of conveyance members such as collets and lead frame feeders accurately and in a short time.

(E. Means for solving the problems) In order to solve the above problems, the bonding apparatus of the present invention has the following steps:
The present invention is characterized in that the position is recognized by a video camera, and the conveyance member is moved by the conveyance movement mechanism by the difference between the recognition result and the target position to perform positioning.

(F. Effect) According to the bonding apparatus of the present invention, the position of conveying members such as collets and lead frame feeder rails is recognized using a video camera, and if there is a deviation from the target position, the deviation is automatically corrected. Since the positioning is performed using the following steps, accurate and quick positioning can be performed automatically.

(G. Embodiment) [FIGS. 1 to 7] The bonding apparatus of the present invention will be described in detail below according to the illustrated embodiment.

(a, First Embodiment) FIGS. 1 to 4] FIGS. 1 to 4 show a collet mounting mechanism according to the first embodiment of the present invention, and FIG. It is a perspective view showing the entire attachment mechanism, and the attachment mechanism will first be explained according to this figure.

1 is a base, 2.2 is a pair of support pieces provided at both ends of one side of the base 1, and one end of the Y-tilt plate 4 is rotatable between the support pieces 2.2 about a shaft 3.3. is supported by Reference numeral 5 denotes a Y-tilting mechanism, which rotates the Y-tilting plate 4 about the axis 3.3 by moving the rotation end of the Y-tilting plate 4 up and down by a Y-tilting motor 6, which is attached to the base 1 and does not appear in FIG. make it move. The principle of this Y tilt mechanism 5 is the same as that of the X tilt mechanism described later, and will not be explained here.

7.7 is a support piece provided at both ends of one side of the Y-tilt plate 4 (one side perpendicular to one side of the base 1 on which the support piece 2.2 is provided) (However, one side 7 does not appear in the drawing). One end of the X-tilt plate 9 is rotatably supported via a shaft 8.8 (the other 8 does not appear in the drawings).

Reference numeral 10 denotes an X-tilting mechanism attached to the Y-tilting plate 4, in which a motor 12 is attached to a U-shaped motor holding bracket 11 with the rotational axis facing vertically, and the upper and lower pieces 13 are raised and lowered by the motor 12. ing. The upper and lower pieces 13 support the rotating end of the X-tilt plate 9 from below,
By being raised and lowered by the motor 12, the Y-tilt plate 9 can be rotated about the shaft 8.8.

Reference numeral 14 denotes an X moving mechanism attached to the X tilting plate 9, which is configured to move the X moving table 16 in the X direction by a motor 15. A guide 17 supports the motor 15 and guides the X-movement table 16 in the X direction.

Reference numeral 18 denotes a Z moving mechanism attached to the X moving table 16, which allows the lifting block 2] to be moved in the vertical direction (two directions) by a two-movement motor 20 attached to a U-shaped bracket 19.

Reference numeral 22 denotes a Y-moving mechanism attached to the lifting block 21, which moves the Y-direction moving flock 27 in the Y-direction by rotating a rotary screw 25 rotatably supported in the Y-direction by a U-shaped bracket 24 with a motor 26. can be moved to A bonding arm 28 is fixed to the block 27. A collet attachment part 29 is fixed to the tip of the arm 28, to which a collet (34) to be described later is replaceably attached. 30
is a pulley for rotating the collet in the rotational direction, 3
1 is the same (belt), 32 is the same (motor), and plays the role of correcting the deviation in the rotation direction (θ direction) of the collet.

Reference numeral 33 denotes an engaging piece for fixing the collet 34 to the collet mounting part 29. The intermediate part is supported by a rotating bearing pressure via a shaft 36 by a pair of support pieces 35 and 35 provided on the collet mounting part 29, and the tip The collet 34 can be fixed to the collet attachment part 29 by engaging with the engagement recess 34a of the collet 34 with the engagement claw.

37 is a replacement collet stand on which the replacement collet is placed, and has a plurality of recesses 38, 38, and 38 on the top surface for holding the collet.
... are formed, and eject bins 39, 39, 39, 39 are provided corresponding to the recesses 38, 38, . . . so as to be movable up and down in the vertical direction. The eject pins 39, 39, 39, 39 rotate in a direction to disengage the engagement piece 33 of the collet mounting portion 29, which is located above the exchange collet stand 37 and faces the collet exchange. play a role. Reference numeral 40 denotes a fork for preventing the old collet from following the old collet 34, which causes the old collet 34 to separate from the collet mounting part 29 when the old collet 34 is left in the recess 38 for holding the collet.

41 is a video camera for collet position detection;
4 from directly below and from the four outer edges 34b, 34b, 34b, 34b of the collet 34 in the X direction and Y direction.
Recognize the direction and position in the θ direction (rotation direction). Then, the inclination of the collet 34 in the X direction and the Y direction is recognized from the focus on the four side edges of the lower end of the collet 34.

The position of the video camera 41 is fixed, and its central axis is taken as the reference position of the center of the collet 34.

FIG. 2 is a cross-sectional view showing the state of the collet mounting portion and the replacement collet stand at the time of collet replacement.

The collet 34 is rotatably attached directly to the collet fitting hole 42a of the rotating member 42, and the rotating member 42 is rotatably attached to the collet mounting portion 29 by a bearing 43. And motor 3
When 2 rotates, the rotation causes the belt 31 and pulley 30
The collet 34 is transmitted to the rotating member 42 via the .theta. direction, and the collet 34 is positioned in the .theta. direction. 44 is a tube for vacuum suction. The tube 44 has appropriate elasticity, and even when the motor 32 rotates, the collet rotation angle required to adjust the square collet 34 in the θ direction is limited to a maximum of 90 degrees as shown in FIG. Since it is a degree, there is no risk that the tube 44 will be twisted to the extent that it cannot handle it.

FIG. 4 is a block diagram showing a control system of the mechanism shown in FIG. 1.

First, a plurality of different collets 34, 34, .
．． 38. Leave it on top. Then, the collet mounting part 29 is positioned on the predetermined collet 34 of the collet stand 37 by the action of each of the x, y, and z transfer iso-yoko 14, 22, 18, and the locking piece 33 is opened by the eject bin 39. while lowering the collet mounting part 29 and attaching the collet 3.
4 into the collet fitting hole 42a of the rotating member 42,
When the fitting is completed, the collet mounting portion 29 is raised and the eject bin 39 is retracted to engage the locking piece 33 with the engagement groove 34a formed on the outer periphery of the collet 34, thereby setting the collet 34. Finish.

Next, the collet mounting portion 29 is moved to the original position directly above the video camera 41.

Next, a video camera 41 images the collet 34 from directly below. And four side edges (edges) 34b, 34b, 3
4b, 34b is recognized after adjusting the position in the Z-axis direction until one of the side edges 34b is in focus. Specifically, first, when one of the pair of opposing side edges 34b, 34b is out of focus, it can be said that the tilt is not in focus, so a tilt motor, for example 12, is driven to eliminate the tilt. Next, regarding the remaining pair of opposing side edges 34b, 34b, if the focus is not on one side edge 34b, it can be said that the inclinations do not match. ) to adjust its inclination.

In this way, when the focus is completely achieved, positioning is performed in the X direction, Y direction, and θ direction.

In FIG. 4, numeral 45 is a calculation means that forms the core of the control system. The collet position is recognized by the collet position recognition system 46 by processing the signal obtained by imaging with the video camera 41, the collet position is calculated by the collet position calculation system 47, and the deviation from the predetermined reference position is calculated by the correction value calculation system. 48, and the calculation means 45 calculates the correction value necessary to eliminate the deviation.
Give feedback. Then, the calculation means 45 controls a pulse generator 49 that generates a control pulse for position correction, and the pulse generator 49 is controlled by the calculation means 45 and sends a control pulse to the motor driver 5o. When the motor driver 50 receives the control pulse, it drives each pulse motor 6, 12, 15, 20, 26, 32 based on the control pulse to align the collet. After that, a bullet bonding operation is performed.

According to such a collet mounting mechanism, the position of the collet 34 is recognized by taking an image with the video camera 41, the error from a predetermined reference position is determined, and the position of each motor 6.1 is determined.
2.15.20.26.32 drive in the X direction, Y direction, θ direction, X tilt direction,
Since positioning in the Y-tilt direction is automatically performed, setup for collet replacement can be performed accurately and quickly, and moreover, labor can be saved.

Note that 51 is a memory for storing data related to calculations, etc.;
52 is a fork drive system, and 53 is an injection bin drive system, which are controlled by the calculation means 45.

By the way, each of the above motors 6.12.15.20.26.
32 is a pulse motor, but it goes without saying that a servo motor may also be used.

(b, Second Embodiment) [Figures 5 to 7] Figures 5 to 7 show a lead frame feeder width changing mechanism which is a second embodiment of the present invention. The figure is a perspective view showing the main parts, and FIG. 6 is a sectional view, and the structure will be explained with reference to both figures.

54.54 is a guide that guides a pair of parallel rails 55.55 in a direction perpendicular thereto (however, one guide 54 does not appear in the drawing);
．． 55 constitutes a lead frame feeder 56 for conveying the lead frame 57. 58.5
Reference numeral 8 denotes a threaded piece provided at the center of the lower surface of the middle part of the rail 55.55, and the threaded piece 58.58 is provided at the center of the rail 5
5.55 are formed, and the screw hole 59 of one of the screwing pieces 58 has a reverse thread.

Reference numeral 60 denotes a rotating shaft, and a threaded groove 61.61 is formed on the outer peripheral surface of the rotating shaft to be threaded into the threaded hole 59.59. Then, one thread groove 61 is reversely threaded, and the above-mentioned reverse threaded screw hole 59
It is screwed together. Therefore, by rotating the rotating shaft 60 in one direction, the pair of rails 55.55 are moved closer to each other (in the direction), and by rotating the rotating shaft 60 in the opposite direction, the pair of rails 55.55 are moved away from each other. Be able to do it.

62 is a pulley 63 fixed to the rotating shaft 60, and a pulley 63 is fixed to the rotating shaft of the pulley 62 and the motor 64.
This is a belt suspended from 5. The distance between the rails 55 can be increased or decreased by rotating the motor 64. Note that 66 and 66 are limit sensors, and 67 is an origin sensor.

68.68 is a target piece formed on the outer surface of the rail 55.55, and a target mark 69.69 serving as a position detection index is provided on the upper surface of the target piece.

70 is a video camera for detecting the feeder position and lead frame position, and is supported by an X-Y drive mechanism 71 so as to be movable in the X direction and the Y direction. 72 is a pillar for supporting the video camera 70, and 73 is the same (support arm).

28 is a bullet bonding arm, which is indicated by a two-dot chain line.

FIG. 7 is a block diagram showing the control system of the lead frame feeder width changing mechanism.

In the figure, 74 is a calculation means that forms the core of the control system;
Reference numeral 5 denotes a data memory for storing data related to the feeder position, video camera position, etc. required for calculation; 76 a pulse generator for generating control pulses for driving the motor 64; and 77, the pulse generator 76. This is a motor driver that drives the motor 64 controlled by the motor 64.

78 is a pulse generator that generates control pulses for driving the X motor 80 and Y motor 81 (not shown in FIGS. 5 and 6) of the camera XY drive mechanism 71, and 79 is a pulse generator that generates the pulses. motor 80.81 controlled by device 78
This is a motor driver that drives the.

Reference numeral 82 denotes a target position recognition system that processes signals obtained by imaging with the video camera 70 to recognize two targets 69 and 69, and 83 a feeder position calculation system that calculates the position of the feeder from the recognition result of the recognition system 82. , 84 is a calculation system that calculates a correction value from the feeder position calculation result, and this calculated correction value is fed back to the calculation means 74.

Next, the operation will be explained.

Information regarding feeder positions for each type of lead frame is stored in memory 75 in advance. When changing the lead frame, the feeder 5 is moved by the motor 64.
6 to the standard value. Specifically, the right rail 55 in FIG. 6 is opened to the position detected by the origin sensor 67, and then the motor 64 is rotated by a predetermined number of pulses based on information regarding the feeder position stored in advance.

Then, the interval between the feeders 56 becomes a value that is relatively close to the interval corresponding to the width of the new lead frame, but it is normal that errors occur due to backlash, backlash, etc.

Next, the video camera 70 captures an image of the target mark 69 on one rail 55 from directly above in a state of just focus, and then the X-Y drive mechanism 71 moves the video camera 7o to the target mark 69 on the other rail 55.
from directly above in the Y direction to a position where the image can be captured in just focus. In this process, the accurate spacing between the rails 55 is calculated, a correction value is calculated from the difference between the calculation result and the target value, and this is fed back to the calculation means 73. Then, the motor 64 is controlled to rotate in the direction in which the error is eliminated, and the distance between the rails 55 becomes a predetermined value.

The video camera 70 is also used during bullet bonding to recognize the lead frame and align the collet attached to the tip of the bonding arm with respect to the lead frame.

According to the feeder width changing mechanism described above, the width of the lead frame feeder 56 is accurately measured using the video camera 70, and if there is an error, it is automatically corrected, so that the feeder width can be changed accurately, resulting in labor savings. be able to.

Incidentally, each motor 64, 80, 81 may be a pulse motor, or may be a servo motor.

(H, Effects of the Invention) As described above, the bonding apparatus of the present invention recognizes the position of conveyance members such as collets and lead frame feeders using a video camera, and adjusts the conveyance by the difference between the recognition result and the target position. The present invention is characterized in that the conveying member is moved and positioned by a moving mechanism.

Therefore, according to the bonding apparatus of the present invention, the position of the rail of the conveying member such as the collet or lead frame feeder is recognized by the video camera, and if there is a deviation from the target position, the deviation is automatically corrected and the position is determined. Therefore, accurate and quick positioning can be performed automatically.

[Brief explanation of the drawing]

Figures 1 to 4 are for explaining a collet mounting mechanism which is one embodiment of the bonding device of the present invention.
Figure 1 is an overall perspective view, Figure 2 is a sectional view showing the state of the collet mounting part and replacement collet stand during collet replacement, Figure 3 is a bottom view of the collet, and Figure 4 is a block diagram showing the control system. 5 to 7 are for explaining a feeder width changing mechanism which is a second embodiment of the present invention.
FIG. 5 is a perspective view showing the main parts, FIG. 6 is a sectional view, FIG. 7 is a block diagram showing the control system, and FIGS. 8 and 9 are sectional views showing different conventional examples. Explanation of symbols 34...Collet, 41...Video camera, 56 (55)...Lead frame feeder, 70...Video camera. Video camera 5:5 control elA thread front 11. Figure 7

Claims (3)

[Claims] (1) Detect the position of the transport member using a video camera, and position the transport member by moving the transport member by the error between the detection result and the target position using a transport member moving mechanism so as to eliminate the error. A bonding device characterized by: (2) The bonding apparatus according to claim (1), wherein the conveyance member is a collet for conveying semiconductor pellets, and a video camera is provided to photograph the collet from below. (3) The conveying member is a lead frame feeder consisting of a pair of rails that are parallel to each other and have a variable interval, and a video camera is movably provided so as to photograph the target marks on the two rails. The bonding apparatus according to claim (1), characterized in that: