JP2940524B2 - Wire bonding equipment - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a wire bonding apparatus, and more particularly to a wire bonding apparatus capable of controlling a loop shape of a wire.

[0002]

2. Description of the Related Art A conventional wire bonding apparatus will be described with reference to FIG.

In FIG. 8, reference numeral 1 denotes a frame which is a main part of a head portion, and various members and components described below are integrally attached to the frame 1. Reference numeral 2 denotes a spool which is mounted on the frame 1 and around which a wire 3 such as a gold wire is wound. Reference numeral 4 denotes a motor for rotating the spool 2 to draw out the wire 3. At this time, the wire 3 is led out in the tangential direction N of the spool 2 so that the wire 3 is not twisted. Reference numeral 5 denotes an air tensioner mounted below the frame, and reference numeral 6 denotes a cut clamper provided below the air tensioner 5. A pad 7 attached to the cut clamper 6 opens and closes the pad 7 to clamp the wire 3.

[0004] Reference numeral 8 denotes a capillary mounted below the cut clamper 6, which is held at the tip of a transducer 9 projecting from the frame 1. Reference numeral 10 denotes a torch electrode which approaches the lower end of the wire 3 led out of the capillary 8 and forms a molten ball 11 at the lower end of the wire 3 by electric spark. Reference numeral 12 denotes a camera mounted on the upper portion of the frame 1, and reference numeral 13 denotes a lens barrel thereof, which captures an image of a bonding point below the capillary 8. Reference numeral 14 denotes a lead frame, and reference numeral 15 denotes a semiconductor element (pellet) mounted on the lead frame 14.
The inner lead 14 </ b> A and the electrode (bonding pad) of the semiconductor element 15 are connected by the wire 3.

Reference numeral 16 denotes a spool 2 and an air tensioner 5
Reference numeral 17 denotes a guide for the wire 3 erected on the air outlet 16. By blowing air to the wire 3 with the air blowing section 16, the wire 3 can be floated between the guides 17 while applying tension to the wire 3. Transmission sensors 18 are provided above and below the guide 17,
When the lower transmission type sensor 18 is shielded from light by a wire, the motor 4 rotates and the wire 3 is drawn out from the spool 2. When the upper transmission type sensor 18 is shielded from light, the motor 4 stops and the spool 2 This stops the wire 3 from coming out.

Next, the operation will be described with reference to FIGS. 9 (a) to 9 (d).

As shown in FIG. 9A, the head portion is initially located above the lead frame 14, and at this position the torch electrode 10 is brought close to the lower end of the wire 3 led out of the capillary 8, and the torch electrode 10 A high voltage is applied to 10 to form a molten ball 11 at the lower end of the wire 3.

Next, as shown in FIG. 9B, the capillary 8 is lowered, and the ball 11 lands on the bonding pad on the semiconductor element 15 for bonding. This is generally called first bonding. At this time, the ultrasonic wave transmitted from the transducer 9 causes the capillary 8
Vibrate at a high frequency to promote bonding of the ball 11.

Next, as shown in FIG. 9 (c), the capillary 8 is raised and moved in the horizontal direction at the same timing to form a loop in an arch shape. It is pressed against the inner lead (A) 14A, and the ultrasonic wave transmitted from the transducer 9 is also used for pressure bonding. This is generally called second bonding.

Next, as shown in FIG. 9D, the wire 3 is clamped by the cut
To raise the capillary 8 after cutting the wire 3 from the bonding point. As described above, the bonding pad of the semiconductor element and the inner lead 14A are connected by the wire 3.

Next, the loop shape and the loop control mode will be described with reference to FIGS.

As shown in FIG. 10, the first post-bonding wire 3 rises right above, draws an arc at the peak, and then linearly descends to the second bonding point. If there is no restriction such as a loop height, bonding is performed in this shape. A loop control mode for forming such a loop is called a normal mode.

However, as shown in FIG.
The difference D between the height of the bonding surface of the inner lead 14A and the first bonding point 19 on the semiconductor element 15 and the height of the second bonding point 20 of the lead frame 14 is large.
When the distance E between the first bonding point 19 and the edge of the semiconductor element 15 is longer than the distance L of the semiconductor element 15, the wire portion 3 comes into contact with the edge portion of the semiconductor element 15 by using the loop control in the normal mode. A short defect called a touch may occur.

In order to prevent the edge touch, the wire 3 is connected to the first bonding point 19 as shown in FIG.
, The wire 3 is formed in a loop substantially parallel to the surface of the semiconductor element 15 at the height H, and then formed into a loop shape descending to the second bonding point 20. A loop control mode for forming such a loop is called a trapezoidal mode.

Next, the trajectory of the capillary 8 forming a loop between the normal mode and the trapezoidal mode will be described in detail.

FIG. 13 shows a normal mode capillary 8.
It is a side view which shows the locus | trajectory. After the first bonding, the first
Raising the capillary 8 perpendicularly from the bonding point 19, pull the wire 3 to a height h _1, after moving the R ₁ only capillary 8 (commonly referred to as reverse motion) in a direction opposite to the wiring direction in its height, capillary after raised to the peak height h _p of 8, it is lowered toward the second bonding point 20. By an air tensioner 5 shown in FIG. 8, the wire 3 is always tension applied to be sucked into a capillary 8, when raised to a height h _p as shown in Figure 13, the wire over a length which is bonded Even if 3 is pulled out, it does not sag, and has an arched loop shape as shown in FIG. In addition, the wire is distorted by the reverse motion of the capillary 8, and the wire is bent starting from the wire, thereby forming the rising portion of the wire 3 from the ball 11 and the wiring ring height H.

FIG. 14 is a side view showing the trajectory of the tip of the capillary 8 in the trapezoidal mode. The process is the same as the normal mode up to the point where the reverse motion is performed to form a rising portion of the wire 3 from the ball 11. Then after raising the capillary 8 to h _2, it is moved to the point T with a slight lowered in the Z direction in a direction opposite to the again wiring direction. At this time, the wire 3 becomes hazy as in the case of the reverse motion. The following capillary 8 rises to h _p, to descend to the second bonding point is the same as the normal mode. By giving the wire 3 a habit at the point T, as shown in FIG. 12, a loop shape that is inflected at the T portion and forms a trapezoidal shape is obtained.

In general, in both the normal mode and the trapezoidal mode, when the distance L between the first bonding point 19 and the second bonding point 20 is determined, the trajectory of the capillary 8 is calculated and determined in all of the X, Y, and Z directions by automatic calculation. Has become. However, although the trajectory is determined as a numerical value by the above-mentioned automatic calculation, a difference appears in the loop shape due to the machine difference of the bonding apparatus. Therefore, a parameter for correcting the loop shape is provided for each bonding apparatus.

FIG. 15 shows parameters for correcting the loop shape of the normal mode related to the present invention.
This will be described with reference to FIG. The parameters for correcting the loop shape include a loop height correction 21, a reverse height correction 22, a reverse amount correction 23, a Z timing correction 24, and a Z speed correction 25. The loop height correction 21 is shown in FIG.
As shown in the figure, when the value of + is entered, the loop height becomes higher, and the overall loop shape becomes inverted in the direction opposite to the wiring direction. When the value of-is input, the loop height becomes lower and the wiring direction becomes smaller. It becomes a looped shape of a cramped fall.

The reverse height correction 22 corrects the height of the reverse. As shown in FIG. 16B, when a value of + is input, a loop shape of the cat is seen, and a value of-is input. And before falling. The reverse amount correction 23 is
It corrects the horizontal movement amount of the reverse motion.
As shown in FIG. 6 (c), when a value of + is entered, the loop height is increased, and when a value of-is entered, the loop height is decreased, resulting in a tightness.

The Z timing correction 24 corrects the timing at which the capillary 8 starts to descend toward the second bonding point 20 while moving in the wiring direction from the peak height, as shown in FIG. To
When the value of + is entered, the loop height becomes high, and the shape becomes looped in the opposite direction to the wiring direction, and the shape of the cat is turned upside down. When the value of-is entered, the loop height becomes low and becomes toward the wiring direction. It becomes a looped shape of a collapsed flat.

The Z speed correction 25 is for correcting the speed of descending toward the second bonding point 20. When a value of + is inserted as shown in FIG. 16 (e), the loop height decreases, and When the value of-is entered, the loop height becomes high, and the shape becomes a cat's back.

A wire bonding apparatus for controlling edge length by controlling a loop length is disclosed in Japanese Patent Laid-Open No. Hei 6-17719.
No. 5 publication. In this apparatus, as shown in FIG. 17, the bonding step D and the loop length of the bonding wire 3 are calculated from the positions of the bonding pads on the semiconductor element 15 recognized by the recognition unit and the bonding positions of the inner leads 14A on the lead frame 14. 1 of the operation portion, for moving upward the capillary from the position of the bonding pad height at which the reverse height, moving the capillary 8 which moves the reverse to a height h ₁ on the opposite side of the bonding locations When the capillary 8 moved by the reverse amount R ₁ and the capillary 8 moved by the reverse amount R ₁ to the highest position, the amount of elevation of the capillary 8 which is the height of the upper surface of the semiconductor element 15 is: A second calculated from the bonding step D and the loop length It is characterized by having a calculation unit.

As shown in FIG. 18, the wire bonder described in Japanese Patent Application Laid-Open No. 6-132347 is based on the mounting position of the semiconductor element 15 recognized by the first recognition unit and the design data of the mounting position. A first calculating unit for calculating the displacement of the mounting position, and calculating a position of the bonding pad in the semiconductor element 15 from the displacement of the mounting position; and a position of the lead frame 14 recognized by the second recognition unit. A second calculating unit that calculates the position shift from the position design data, and calculates a bond position of the inner lead in the lead frame 14 from the position shift, a position of the bonding pad, the bonding position, From the design data of the semiconductor device, the bond of the portion located above the semiconductor element 15 is determined. A third calculation unit length of Inguwaiya 3 is calculated, the reverse high, the reverse amount, the height the height h _p and the highest position from the upper surface of the semiconductor element 15 to rise after reverse motion The trajectory of the capillary 8 when moving the capillary 8 to the bonding position is determined by the loop length of the bonding wire 3 calculated from the position of the bonding pad and the bonding position, and the position of the portion positioned above the semiconductor element. The length of the bonding wire 3, the design data of the position of the bonding pad and the bonding step D calculated from the design data of the bonding position, and the height of the bonding wire 3 positioned above the bonding pad from the upper surface of the semiconductor element 15. First design data that is height and And an automatic selection unit that is automatically selected based on second design data that is a height of the bonding wire 3 located above a peripheral portion of the semiconductor element 15 from an upper surface of the semiconductor element 15. Features.

[0025]

The first problem is that, in the prior art, when the trapezoidal mode is used to prevent the edge touch, the speed required for bonding one wire (generally, the bonding speed) is reduced. Is reduced).

The reason is that, as described with reference to FIG. 14, the trapezoidal mode requires an operation of adding another habit to the wire after the reverse motion as compared with the normal mode, so that the moving distance of the capillary increases. In particular, since it is necessary to stop the upward movement of the Z-axis and perform the downward movement, the time required for the operation of the capillary becomes long.

The second problem is that, for the same reason as described above, when the trapezoidal mode is used, the length of the wire used for bonding one wire becomes longer and the amount of wire used increases.

The reason is that, in general, in a conventional wire bonding apparatus, quality check including a loop shape is performed using a plurality of semiconductor elements and a lead frame before fully automatic production is started, and mounting of the semiconductor elements is performed. The setting with a margin is performed in consideration of the deviation such as the displacement and the thickness of the mounting agent for mounting the semiconductor element. If the trapezoidal mode is used to prevent the edge touch, the distance between the edge portion of the semiconductor device and the wire becomes longer than necessary.

In the wire bonding apparatus disclosed in JP-A-6-177195, the reverse height, the reverse amount, and the capillary rise amount are calculated to optimize the loop height and the loop shape. Since it is not possible to control the loop shape with a shape close to a trapezoid only by the amount of rise of the capillary, there is a limit in preventing the edge touch.

In the wire bonder disclosed in JP-A-6-132347, the bonding step D is calculated from the design data of the position of the bonding pad on the semiconductor element and the design data of the position of the second bonding point of the inner lead.
The parameters relating to the loop shape are automatically selected based on the stored values. However, as described above, the thickness of the mounting agent for mounting the semiconductor element on the lead frame, the island mounted from the semiconductor element, Considering the submerged processing accuracy and the variation in the thickness of the semiconductor element, the disadvantage of a wire bonder that fixes the bonding step D to a fixed value is that the distance between the edge of the semiconductor element and the wire becomes longer than necessary. In addition, there is an operation of storing the bonding step D in advance for each type change, and in some cases, it is necessary to calculate the bonding step of a product to be bonded.

A first object of the present invention is to provide a wire bonding apparatus capable of preventing edge touch without lowering the bonding speed. A second object is to provide a wire bonding apparatus that can prevent edge touch without stretching a wire longer than necessary.

[0032]

A wire bonding apparatus according to the present invention detects the position and height of a bonding pad of a semiconductor element and an inner lead of a lead frame for each IC to be bonded.
Based on this, bonding parameters relating to the operation of the capillary are automatically calculated, and the operation of the capillary is automatically controlled to perform bonding.

The positions and heights of the bonding pads of the semiconductor element and the inner leads of the lead frame are detected by using the same camera.

Further, in a wire bonding apparatus having a function of continuously recognizing a plurality of semiconductor elements and then bonding the continuously recognized semiconductor elements, a wire bonding apparatus having a function of detecting a second semiconductor element detected earlier. The calculation of the bonding parameter related to the operation of the capillary performed based on the position and the height of the bonding pad is performed while the position and the height of the bonding pad of the third semiconductor element are detected and before the position and the position of the bonding pad are detected. This is performed during the bonding of the first semiconductor element whose height has been detected and the calculation of the bonding parameter has been completed.

[0035]

The position and height of the bonding pad of the semiconductor element and the inner lead of the lead frame are detected for each IC to be bonded, and the operation of the capillary is automatically controlled based on the detected position. The optimal loop shape is obtained.

Further, since the same camera is used for detecting the position and height of the bonding pad of the semiconductor element and the inner lead of the lead frame, the position and height of the semiconductor element and the lead frame immediately before bonding are determined. Can be detected simultaneously.

Further, the determination is performed based on the position and height of the bonding pad of the second semiconductor element detected in advance.
The calculation of the bonding parameter related to the operation of the capillary is performed while the position and height of the bonding pad of the next third semiconductor element are detected or prior to the detection of the position and height of the bonding pad and the calculation of the bonding parameter. Since the bonding is performed during the completed bonding of the first semiconductor element, the time required for calculating the bonding parameters can be offset.

[0038]

Next, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 shows the first embodiment of the present invention.
It is a block diagram for demonstrating embodiment.

Referring to FIG. 1, the preferred embodiment of the present invention comprises a storage unit 26, a recognition unit 27, a first operation unit 28,
It is mainly composed of a second operation unit 29 and a third operation unit 30. The storage unit 26 stores all data stored in the wire bonding apparatus. The storage unit 26 stores data 31 of the distance E between the bonding pad and the edge of the semiconductor device 15 related to the present invention, and the first data 31 on the semiconductor device 15. Bonding point 19 and inner lead 14A of lead frame
The coordinate data (bonding coordinate data) 32 of the upper second bonding point is also stored. The recognition unit 27 includes a camera that captures an image and an image processing device built in the wire bonding device.

The camera used for the recognition unit 27 is a CCD camera 3 with a laser focus displacement meter as shown in FIG.
3 is desirable. The CCD camera 33 with the laser focus displacement meter is divided into a light beam for detecting displacement by a half mirror 35 inside the laser focus displacement meter 34 and a light beam for capturing an image by the CCD camera 36. Image data input from the CCD camera 36 is sent to an image processing device (not shown). Also, displacement data is output from the laser focus displacement meter.

Next, the operation of the embodiment of the present invention will be described in detail with reference to FIGS. The data of the amount of displacement or inclination of the relative position between the semiconductor element 15 and the inner lead detected by the recognition unit 27 and the registered bonding coordinate data 32 are sent to the first arithmetic unit, and the first bonding point 19 and the The distance L between the second bonding points 20 is calculated. The height data of the semiconductor element 15 detected by the recognition unit and the height of each of the bonded parts of the inner leads 14A are sent to the second calculation unit, and the bonding step D is calculated.

The distance L obtained by the first arithmetic unit and the second
Of the bonding step D obtained by the calculation unit of FIG.
Is transmitted to the third calculation unit. In the third calculation unit, parameters relating to the loop shape are calculated from the three types of data sent so that the distance A between the wire 3 and the edge of the semiconductor element 15 becomes an optimum loop shape that minimizes the necessary. You.

The bonding of the capillary 8 is controlled by controlling the operation of the capillary 8 based on the parameters obtained as described above.

[0044]

Next, an example of the first embodiment of the present invention will be described in detail with reference to the drawings. An example is a thin package called an 8-pin TSSOP having a diameter of 5
Description will be made using a product for bonding a 0 μm gold wire.

FIGS. 3A and 3B are a plan view and a side view showing a state in which the semiconductor element 15 is mounted on the island 37 of the lead frame 14 for the 8-pin TSSOP. This state is shown in a state where it is transported to a portion and is sandwiched between a lead frame clamper (not shown) and the heater plate, and the position is fixed.

FIG. 4 is a block diagram showing a configuration of a main part of the wire bonding apparatus according to the embodiment of the present invention. The data 31 of the distance between the bonding pad and the edge of the semiconductor element 15 and the bonding coordinates 32 are input by a floppy disk or a key (not shown) or by teaching while operating the manipulator while looking at the monitor screen. The input data is stored in the random access memory (R) of the wire bonding apparatus.
AM) 37.

The recognition section 27 comprises a CCD camera 33 with a laser focus displacement meter and an image processing device 38.

The first arithmetic unit 28 and the second arithmetic unit 29
The third arithmetic unit 30 is configured in the CPU 39 in the wire bonding apparatus.

Next, the operation of the embodiment of the present invention will be described with reference to FIGS.

First, the image of the recognition point LF1 forming the characteristic pattern of the inner lead 14A of the lead frame 14 is taken by the CCD camera 33 with the laser focus displacement meter, and the height of the surface is measured. In the case of a fine pitch product, in order to increase the position detection accuracy, another point LF at the diagonal of the recognition point LF1 of the inner lead 14A is used.
Some recognize the second lead, but here the inner lead 1
The recognition point of 4A is LF11 only. Next, a CCD camera 33 with a laser focus displacement meter is mounted on an XY (not shown).
It moves by the operation of the table, and measures the height of the surface at two points of the recognition points PE1 and PE2, which are characteristic patterns on the semiconductor element 15, while capturing an image.

The captured image signal 40 is sent to an image processing device 38. Although not described here, the current position is recognized by a pattern matching method, and the relative displacement between the inner lead 14A and the semiconductor element 15 is determined. Tilt is detected,
The data is sent to the first arithmetic unit 28 in the CPU 39. Also, the inner lead 14A of the lead frame 14 and the semiconductor element 15 are detected by the laser focus displacement meter 34, respectively.
Is sent to the second arithmetic unit 29 in the CPU 39.

Next, as described above, in the first arithmetic unit 28, the bonding coordinates 3 stored in the RAM 37 are stored.
The distance L between the first bonding point 19 and the second bonding point 20 is calculated from the data of the relative displacement and the inclination of the semiconductor element 15, the inner lead 14A and the semiconductor element 15,
The second calculating unit 29 calculates the bonding step D.

Data on the calculated distance L, the bonding step D, and the distance E between the bonding pad and the edge of the semiconductor element 15 stored in the RAM 37 are sent to the third arithmetic unit 30. The third calculation unit 30 calculates a loop-shaped parameter based on the three data so that the proximity distance A between the edge of the semiconductor element 15 and the wire 3 is kept constant. In general, the proximity distance A between the edge of the semiconductor element 15 and the wire 3 is set so as to maintain two or more wire diameters. In the present embodiment, since a 50 μm diameter wire is used, it is 100 μm.

Next, the calculation of the parameters of the loop shape will be described with reference to FIG. FIG. 5 is a cross-sectional view taken along the line BB shown in FIG. 3, and shows the distance L between the first bonding point 19 and the second bonding point 20, the bonding step D, and the distance E between the edge of the semiconductor element 15 and the bonding pad. 3 shows the shape of the wire 3 bonded by the parameters of the loop shape calculated by the third arithmetic unit 30.

Instead of using the trapezoidal mode as shown in FIG. 12, the third arithmetic unit 30 keeps the normal mode and the loop shape as shown in FIGS. 15 and 16 so that the distance A becomes about 100 μm. Is calculated so that the parameter for correcting becomes an appropriate value. For example, FIG.
As shown in (d) and (e), the Z timing correction 24
In FIG. 16 (a), the value on the negative side and the value on the positive side in the Z speed correction 25 are inserted, and the loop height H is increased as shown in FIG. The loop shape as shown in FIG. 5 can be obtained by lowering the loop height H by putting a value of − in the loop height correction 21.

As described above, for all the wires 3 to be bonded, the loop shape is optimized and bonding is performed. In addition, bonding is performed for all the semiconductor elements 15 and the inner leads 14A of the lead frame 14 while detecting a bonding step D that is different depending on the amount of displacement of the mounting position, the thickness of the mounting agent, and the like. it can.

Since the trapezoidal mode is not used here,
Bonding can be performed without reducing the bonding speed.
When bonding the product having the distance L of about 1.5 mm as in the present embodiment using the trapezoidal mode, the bonding speed is about 0.25 seconds, whereas when bonding in the normal mode as in the present embodiment, , About 0.18 seconds.

The length of the wire 3 in the case of bonding the product as in this embodiment using the trapezoidal mode is about 2.0 mm, whereas in this embodiment, the distance A is minimized by using the normal mode. , The wire 3 can be bonded with a length of about 1.8 mm.

Next, a second embodiment of the present invention will be described.
This will be described with reference to FIGS.

FIG. 6 is a plan view showing a lead frame in which one IC of the lead frame 14 shown in FIG. 3 is formed in a plurality of columns in a vertical direction. Here, the lead frame 14 is formed in four columns. . Generally, the lead frame 14 in which a plurality of rows of ICs are formed vertically is called a matrix lead frame 42.

FIG. 7 shows a matrix lead frame 4 as shown in FIG. 6 using the wire bonding apparatus of the present invention.
2 shows a timing chart when bonding 2 is performed.

Next, the operation of the second embodiment will be described. As described above, the LF1, PE1, and PE2 described with reference to FIG. 3 are recognized as the recognition of the first IC, and after detecting the position data and the height data, the detected data and the data stored in the storage unit 26 in advance are used. An operation for obtaining an optimum loop shape is performed by the first to third operation units. Also, during the calculation of the loop shape, the recognition unit 26 moves by the operation of the XY table (not shown) and
Start recognition of the IC. After the recognition up to the fourth IC is completed as described above, bonding is performed in order from the first IC for which the calculation for obtaining the optimum loop shape has been completed. Also, during the bonding of the first IC, the calculation of the loop shape optimization of the third IC and the fourth IC is performed.

As described above, the matrix frame 42
On the other hand, a method called multi-bonding, in which recognition is continuously performed for a plurality of rows of ICs in advance and then continuous bonding, and an operation for optimizing a loop shape are performed by recognizing the next IC. The bonding time can be offset by performing the calculation during the bonding of the IC for which the calculation of the loop shape has been completed or the IC for which the calculation of the loop shape has been completed. You can bring out all the effects of speeding up.

[0064]

The first effect is that an optimum loop shape free from edge touch can be obtained without lowering the bonding speed. Thereby, the production capacity of the wire bonding apparatus can be brought out.

The reason is that the individual I
This is because a calculation unit is provided for detecting the position and height of each of the semiconductor element and the inner lead of the lead frame for C, and calculating the parameters based on the positions and heights so that bonding can be performed with the normal mode loop control.

The second effect is that an optimum loop shape without edge touch can be obtained without making the length of the wire longer than necessary. Thereby, the wire cost can be reduced.

The reason is that the distance between the wire and the edge of the semiconductor element can be kept to a necessary minimum in the normal mode without using the trapezoidal mode.

[Brief description of the drawings]

FIG. 1 is a block diagram for explaining a first embodiment of the present invention.

FIG. 2 is a sectional view showing a configuration of a CCD camera with a laser focus displacement meter.

FIG. 3 is a plan view and a side view showing a configuration of a semiconductor element and a lead frame according to an embodiment of the present invention.

FIG. 4 is a block diagram showing the configuration of an embodiment of the present invention.

FIG. 5 is a sectional view showing a loop shape according to the embodiment of the present invention.

FIG. 6 is a plan view showing a lead frame for explaining a second embodiment of the present invention.

FIG. 7 is a timing chart showing an operation of a wire bonding apparatus for explaining a second embodiment of the present invention.

FIG. 8 is a side view showing a head portion of a conventional wire bonding apparatus.

FIG. 9 is a side view showing the outline of the loop forming operation of wire bonding.

FIG. 10 is a cross-sectional view showing a loop shape when bonding is performed in a normal mode according to the related art.

FIG. 11 is a cross-sectional view showing a loop shape in a state where an edge touch has occurred by bonding in a normal mode according to the related art.

FIG. 12 is a cross-sectional view showing a loop shape when bonding is performed in a trapezoidal mode according to the related art.

FIG. 13 is a side view showing a trajectory of a conventional capillary in a normal mode.

FIG. 14 is a side view showing a trajectory of a capillary in a trapezoidal mode according to the related art.

FIG. 15 is a side view showing a normal mode loop shape correction parameter according to the related art.

FIG. 16 is a side view showing a tendency of correction of a loop shape when a value of + or − is input to each of the loop shape correction parameters shown in FIG. 5;

FIG. 17 is a block diagram showing a configuration of a conventional example.

FIG. 18 is a block diagram showing a configuration of another conventional example.

[Explanation of symbols]

Reference Signs List 1 frame 2 spool 3 wire 4 motor 5 air tensioner 6 cut clamper 7 pad 8 capillary 9 transducer 10 torch electrode 11 ball 12 camera 13 lens barrel 14 lead frame 14A inner lead 15 semiconductor element 16 air blowing section 17 guide 18 transmission type sensor 19 First bonding point 20 Second bonding point 21 Loop height correction 22 Reverse height correction 23 Reverse amount correction 24 Z timing correction 25 Z speed correction 26 Storage unit 27 Recognition unit 28 First calculation unit 29 Second calculation unit 30 Third arithmetic unit 31 Data of distance E between bonding pad and edge of semiconductor element 32 Bonding coordinates 33 CCD camera with laser focus displacement meter 34 Laser focus displacement meter 35 Half-mi Over 36 CCD camera 37 RAM 38 the image processing apparatus 39 CPU 40 image signals 41 height data 42 Matrix lead frame

Claims (3)

(57) [Claims] 1. A storage unit for storing a distance between a bonding pad of a semiconductor element and an edge of the semiconductor element and bonding coordinates, and a recognition unit for recognizing a shift amount and a tilt of a relative position between the semiconductor element and an inner lead of a lead frame. A first calculating unit for calculating a distance between a first bonding point and a second bonding point based on data between the storage unit and the recognition unit; and the semiconductor element and the inner lead based on data from the recognition unit. A second calculating unit for calculating the step of the bonded portion, data of the distance between the bonding pad of the storage unit and the edge of the semiconductor element, and data from the first calculating unit and the second calculating unit. A third operation unit for calculating an operation parameter of the capillary regarding a loop shape for minimizing the distance between the wire and the edge of the semiconductor element. A wire bonding apparatus. 2. The wire bonding apparatus according to claim 1, wherein the recognition unit includes a camera that simultaneously detects the position and height of the bonding pad of the semiconductor element and the inner lead of the lead frame. And a third arithmetic unit configured to perform wire bonding between the second semiconductor element and the inner lead while wire bonding is performed between the bonding pad of the first semiconductor element and the inner lead of the lead frame. The wire bonding apparatus according to claim 1, wherein the apparatus is configured to calculate an operation parameter of the capillary regarding the loop shape of the wire.