TW202101099A - Mounting device and manufacturing method of semiconductor device - Google Patents

Abstract

The invention provides a mounting device capable of reducing vibration influence in stroboscopic exposure identification shooting and a manufacturing method of a semiconductor device. This mounting device is provided with: an imaging device that moves relative to an object to be imaged and images the object to be imaged; and an illumination device that irradiates light to the subject to be imaged. The illumination device is configured to perform strobe light emission for a plurality of times at a period equal to or less than 1/2 of a vibration period within an exposure time of the imaging device.

Description

Assembly device and manufacturing method of semiconductor device

The assembly device of the present disclosure can be applied to an assembly device using strobe lighting, for example.

In the electronic component assembly device, in order to assemble electronic components with high precision, while the electronic component is sucked and held by the suction nozzle and moved to the assembly assembly part, the electronic component is photographed by a digital camera with an imaging element such as CCD or CMOS. For parts, the captured image information is transformed into electronic signals, and image processing is performed to obtain measured values. Then, based on the measured value, the assembly position is corrected during the movement of the electronic component, and the electronic component is assembled on the printed circuit board. In order to obtain high-precision images with less error when shooting electronic parts with this camera, it is necessary to use a lighting device with a light intensity above the specified level. Generally speaking, flashes caused by strobe lights or continuous flashes caused by halogen lamps are used. Means such as lighting (for example, Patent Document 1). In the exposure (strobe light exposure) of the recognition camera that uses strobe light illumination, by setting the exposure time to a short time of tens of μs, even high-speed actions can be taken, and recognition processing can be performed. [Prior Technical Literature] [Patent Literature]

[Patent Document 1] JP 2000-124683 A

[The problem to be solved by the invention]

In a device where multiple axes are operating at the same time, when recognizing the actions of other axes during shooting, due to the effect of motion vibration, the captured image may have a positional deviation between the person in the static position and the person who is vibrating . In strobe light exposure, for high-speed processing, the electronic circuit composition is mainly based on high light intensity in a short time. The extension of the lighting time of the strobe light has the limitation of the longest light-emitting time, and it is difficult to make the light-emission exceed the inherent device. The imaging is performed in one cycle of vibration (for example, tens of Hz, tens of ms, etc.). The problem to be solved in the present disclosure is to provide an assembly device that can reduce the impact of vibration in the strobe light exposure identification shooting. Other unsolved problems and novel features will be clear from the description of this specification and the accompanying drawings. [Technical means to solve the problem]

A brief description of the outline of the representative persons in this disclosure is as follows. That is, the assembling device includes an imaging device that relatively moves the object to be photographed and photographs the object, and an illuminating device that irradiates the object to be photographed. The lighting device is configured to emit a strobe light multiple times in a period of 1/2 or less of the vibration period during the exposure time of the imaging device. [Effect of Invention]

According to the above assembly device, the influence of vibration can be reduced during strobe light exposure.

The following uses the drawings to describe the embodiment and examples. However, in the following description, the same components are marked with the same symbols and repeated descriptions may be omitted. In addition, in order to make the description clearer, the width, thickness, and shape of each part may be represented in a model compared to the actual state, but it is only an example and does not limit the interpretation of the present invention.

<The first embodiment> Use Figures 1 to 5 to illustrate the shooting system of strobe light exposure. FIG. 1 is a schematic diagram of the structure of the imaging system of the first embodiment. FIG. 2 is a schematic diagram of the action time point of the shooting system of FIG. 1 in the case of short-term shooting after the axis movement is stopped. FIG. 3 is a schematic diagram of light-emitting time points of the photographing system of FIG. 1. FIG. 4 is a schematic diagram of the action time point in the case of setting up a recognition waiting time before the start of recognition and shooting from the stop of the axis motion of the photographing system of FIG. 1. FIG. 5 is a schematic diagram of the action time point of the shooting system of FIG. 1 when the axis action of the shooting system is stopped and shooting multiple times.

As shown in FIG. 1, a photographing system 100S for strobe light exposure includes a controller 101, strobe light illumination power supplies 107a, 107b, a lighting device 103, a photographing device 104 having a camera 104a and a lens 104b, and an identification processing device 105.

The controller 101 outputs the shooting trigger signal to the camera 104a, and the camera 104a receives the shooting trigger signal to start exposure, and at the same time outputs the lighting trigger signal (strobe light trigger) to the strobe lighting power supply 107a, 107b to make the lighting from Starting from the start of exposure, the lamp will be turned on until the set light-emitting delay time. The strobe lighting power supplies 107a and 107b receive the lighting trigger signal from the camera 104a, and supply the lighting current to the lighting device 103. The lighting device 103 is composed of pseudo-coaxial lighting 103a, side lighting 103b, and the like. The illuminating device 103 is to arrange a plurality of light-emitting diodes neatly in a matrix, and to illuminate the workpiece 108 of the photographing object by the light source of the surface light.

As shown in FIG. 2, the illuminating device 103 lights up the strobe light caused by a single light after the light-emitting delay time (Td4) from the shooting trigger signal. The controller 101 inputs a shooting trigger signal to the camera 104a. After the exposure delay time (Td1), the camera 104a opens the shutter to expose the set exposure time (Tex1), and shoots the workpiece 108 of the object through the lens 104b. After the shooting of the camera 104a is completed (after the exposure is completed, the waiting time (Td2) is passed), the camera 104a transmits the image data to the recognition processing device 105. After the waiting time (Td3) is completed, the recognition processing device 105 calculates based on the image data Position deviation etc.

If the time from when the shooting trigger signal is issued to when the recognition calculation is completed is set as the recognition time (Tr), then Tr becomes the following formula. Tr=Td1＋Tex1＋Td2＋Ttr＋Td3＋Trc Here, Ttr is the image transfer time, and Trc is the identification calculation time.

As shown in Figure 3, the time the camera shutter is opening is the exposure time (Tex1), which is Tex1=Td4+Tem+Tm. Here, Tem is the light emission time, and Tm is the exposure margin time. The luminescence delay time, luminescence time, and luminescence intensity are variable values, and the exposure margin time is a fixed value, for example, the following values. Luminous time = 5～107.4μs (0.2μs scale) Luminescence delay time = 10～112.4μs (0.2μs scale) Exposure margin time = 5μs (fixed value) Luminous intensity (light quantity) = 512 steps Strobe light exposure, the exposure time (Tex1) can be set to as short as 20μs, and the actual exposure time, that is, the light-emitting time, can also be set to as short as 10μs, even for moving identification shooting. Recognition, and can achieve high speed. Here, the so-called recognition shooting during movement refers to the case where the imaging device 104 is stationary and the object 108 is moving, the object 108 is stationary and the imaging device 104 is moving, the imaging device 104 Recognition and shooting when both the workpiece 108 and the object are moving.

As mentioned above, in a device where multiple axes are operating at the same time, when recognizing the movement of other axes during shooting, due to the influence of motion vibration, the captured image may be in a static position or when it is vibrating. There will be positional deviation.

If you shoot within a short period of time after the axis motion stops, the recognition error caused by vibration will occur. As shown in Figure 2, the average value of the displacement during exposure is the identification position (the black circle in Figure 2), and the displacement in the negative direction in Figure 2 has an identification deviation. In other words, if shooting with a single strobe light during vibration, the position may shift.

As shown in FIG. 3, by setting the actual exposure time, that is, the light emission time (Tem) to be longer than the vibration period, the shaking caused by the vibration in the captured image will be averaged, and the influence of the vibration can be reduced. For example, when the light-emitting time of the strobe lamp is about 10ms and the vibration period is less than 10ms (the vibration frequency is more than 100Hz), the impact of vibration can be reduced. However, in moving exposure and shooting, in order to increase the illuminance of the strobe light, the output of the lighting power supply becomes high current, and the power supply capacity is insufficient, so it cannot emit light for a long time. If the light-emitting time is as long as 100μs and the vibration period is less than 100μs (the vibration frequency is above 10kHz), the impact of vibration can be reduced.

As shown in Fig. 4, it is conceivable to set up a recognition waiting time (Trw) from the stop of the axis motion to the start of recognition, and wait for the vibration to subside before exposure. In this way, even if the vibration period is long, the impact of vibration can be reduced. However, considering the long vibration period, the recognition waiting time (Trw) must be set longer, and the recognition will take time.

As shown in FIG. 5, it is also conceivable that after the axis motion is stopped, a plurality of times (N times) of shooting and recognition calculations are performed, and their averaging processing is performed. In this way, even if a long identification waiting time is not established, the impact of vibration can still be reduced. However, the identification time (Tr2) will be nearly N times the identification time (Tr) in the situations of Figures 2 and 3, and identification will take time. In the case of Figure 5, it becomes Tr2=N×Tr－(N－1)Td4. Here, N=2.

The stroboscopic light exposure of the above-mentioned first embodiment only emits light once during the exposure time, so it is hereinafter referred to as stroboscopic light exposure once.

<The second embodiment> The shooting system of the second embodiment of the strobe light exposure will be explained using Figs. 6-8. Fig. 6 is a schematic diagram of the configuration of the imaging system of the second embodiment. FIG. 7 is a schematic diagram of the operation time point of the photographing system of FIG. 6. Fig. 8 is a schematic diagram of the light-emitting time point of the photographing system of Fig. 6, Fig. 8(a) is a case where the luminous intensity is set to 1/N, and Fig. 8(b) is a case where the light-emitting time is set to 1/N.

As shown in FIG. 6, the configuration of the imaging system 100 of the second embodiment is the same as that of the imaging system 100S of FIG.

As shown in FIGS. 6 and 7, the controller 101 outputs a shooting trigger signal to the camera 104a. The camera 104a starts exposure after the exposure delay time (Td1) from the shooting trigger signal, and outputs the continuous illumination trigger signal (strobe light trigger) to the strobe light after the set light-emitting delay time (Td4) from the start of the exposure Illumination power supply 107a, 107b. The stroboscopic lighting power supplies 107a, 107b, calculated from the lighting trigger signal, use the preset lighting delay time (Td4), lighting time (Tem), light quantity (luminous intensity), and lighting period (Tep) to make the lighting device 103 emit light The diode emits light. The camera 104a exposes the shutter by opening the shutter for the set exposure time (Tex2), and photographs the workpiece 108 of the subject through the lens 104b. After the shooting of the camera 104a is completed (after the exposure is completed, the waiting time (Td2) is passed), the camera 104a transmits the image data to the recognition processing device 105. After the waiting time (Td3) is completed, the recognition processing device 105 calculates based on the image data Position deviation etc. The camera 104a outputs a plurality of trigger signals for an input of a shooting trigger signal.

The recognition time (Tr3) is the same as in the first embodiment and is as follows. Tr3=Td1＋Tex2＋Td2＋Ttr＋Td3＋Trc However, the exposure time (Tex2) of the second embodiment is longer than the exposure time (Tex1) of the first embodiment, so the recognition time (Tr3) of the second embodiment will be longer than the recognition time (Tr) of the first embodiment The length is exactly (Tex2-Tex1).

As shown in Figure 8, the exposure time (Tex2) is Tex2=Tem×(N-1)+Td4+Tem+Tm. In addition, the light emission period is 10ms or more under the limitation of the current strobe light. For example, if you order Luminous period = 10ms Luminescence delay time = 10μs Luminous time = 50μs Exposure margin time = 5μs N=4 then Tex2=10ms×(4-1)+10μs+50μs+5μs=30.11ms.

The strobe light exposure of the above-mentioned second embodiment emits light several times during the exposure time, so it is referred to as strobe light multiple exposure hereinafter.

In addition, in order to prevent overexposure, the total exposure light intensity of the strobe light multiple exposure of the second embodiment is set to be the same as the light intensity of one exposure of the strobe light of the first embodiment. The amount of light of a strobe light is adjusted by adjusting the luminous time and luminous intensity. For example, as shown in Figure 8(a), when the luminous time of one stroboscopic multiple exposure is set to be the same as the luminous time of one stroboscopic exposure in one exposure time Next, set the luminous intensity of one strobe light multiple exposure to 1/N of the luminous intensity of one strobe light exposure. As shown in Figure 8(b), when the luminous intensity of the first multiple exposure of a strobe light that emits N times in one exposure time is set to be the same as the luminous intensity of one exposure of the strobe light, Set the light emission time to 1/N. In addition, for example, when the light emission time of one strobe light multiple exposure is set to 5 times the light emission time of one strobe light exposure, the light emission intensity of one strobe light multiple exposure is set to the frequency 1/5N of the luminous intensity of one flash exposure.

As shown in Fig. 7, the strobe lamp is turned on multiple times (4 times in this example) within one exposure time. There are multiple exposures of the strobe light in one image, and the vibration components in one image are averaged. For this reason, during the time (exposure time) during which the shutter of the camera is opened, a period of less than 1/2 of the vibration period (Top) is used for multiple strobe light exposures. In this way, the identification deviation caused by vibration caused by shaft motion or other shaft motions, and vibration caused by external disturbances can be reduced.

In the second embodiment, the strobe light is emitted multiple times in one shot, and the multiple exposures are used for averaging, thereby reducing the occurrence of vibration caused by the movement of the axis after the movement axis stops or the movement of the axis other than the recognition target axis. The influence of external disturbances such as the influence of vibration. In addition, the time to start the recognition after the recognition target motion axis stops can be shortened. In addition, only one shooting and data transfer are required, so the transfer time and the processing time for identification calculation can be shortened, and it can be faster than ordinary shooting with multiple strobe lights and one exposure. In addition, with the stroboscopic lighting that can achieve sufficient illumination in a very short time, the actual exposure time (exposure time during the light-emitting period) can be shortened, and it can also be recognized on the move.

<Modifications> The following examples illustrate some representative modifications of the embodiments. In the description of the following modification examples, parts having the same configuration and function as those described in the above-mentioned embodiment may be given the same reference numerals as in the above-mentioned embodiment. In addition, for the description of this part, the description in the above-mentioned embodiment may be appropriately used to the extent that it is not technically contradictory. In addition, a part of the above-mentioned embodiment and all or a part of a plurality of modified examples should be appropriately applied in a complex manner within a range that is not technically contradictory.

(First modification) FIG. 9 is a schematic diagram of the structure of the imaging system of the first modification.

In the second embodiment, the camera 104a generates the trigger signal for multiple times of strobe light emission, but it can also be generated by an external pulse generator of the camera 104a. The camera 104a of the imaging system 100A of the first modification outputs the inputted imaging trigger signal to the pulse wave generating circuit 109, and the pulse wave generating circuit 109 outputs a predetermined number of trigger signals at predetermined intervals based on the inputted imaging trigger signal.

(Second Modification) In the second embodiment, when the luminous time of the first multiple exposure of the strobe light is set to be the same as the luminous time of the first exposure of the strobe light, the luminous intensity of the first multiple exposure of the strobe light is set It is 1/N of the luminous intensity of the first exposure of the strobe light, but the luminous intensity of the first multiple exposure of the strobe light can be set to be the same as the luminous intensity of the single exposure of the strobe light. In this way, in the appearance inspection for foreign object detection, even in the recognition of a darker object with insufficient illumination with one exposure, multiple strobe light exposures can be performed to ensure the brightness for recognition.

(Third modification) It is also possible to combine the imaging system 100S of the first embodiment and the imaging system 100 of the second embodiment. The shooting system of the third modification has the single exposure function of the shooting system 100S strobe light (stroboscopic light single exposure mode) and the shooting system 100 strobe light multiple exposure function (strobe light multiple exposure mode) .

In the shooting system of the third modification example, in the single exposure mode of the strobe light, the camera 104a outputs a single trigger signal based on the shooting trigger signal from the controller 101, and operates like the shooting system 100S. In the strobe light multiple exposure mode, the camera 104a outputs a continuous trigger signal based on the shooting trigger signal from the controller 101, which acts like the shooting system 100. [Example]

Hereinafter, an example of a flip chip bonder (flip chip bonder), which is an example of the assembly device applied to the above-mentioned embodiment, will be described. In addition, flip chip bonders, for example, are used in the manufacture of packages that form a rewiring layer in a wide area beyond the chip area, that is, fan-out wafer level package (FOWLP).

Fig. 10 is a schematic plan view of the flip chip bonder of the embodiment. Fig. 11 is an explanatory diagram of the operation of the pick-up and reversal head and the transfer head when viewed from the direction of arrow A in Fig. 10. Fig. 12 is an explanatory diagram of the operation of the bonding head when viewed from the direction of arrow B in Fig. 10. Fig. 13 is a schematic cross-sectional view of main parts of the crystal grain supply part of Fig. 10. As shown in FIG. 10, the flip chip bonder 10 includes a die supply part 1, a first picking part 2a, a second picking part 2b, a first transfer platform part 3a, a second transfer platform part 3b, and a first bonding part. The part 4a, the second joining part 4b, the conveyance part 5, the board|substrate supply part 6K, the board|substrate delivery part 6H, and the control apparatus 7 which monitors and controls the operation of each part. In addition, the second pick-up portion 2b, the second transfer platform portion 3b, and the second bonding portion 4b are connected to the first pick-up portion 2a, the first pick-up portion 2a, and the first pick-up portion 2a and the first joint portion 4b. The transmission platform portion 3a and the first joining portion 4a are arranged mirror-symmetrically, have the same structure, and operate in the same manner. The symbol "b" of each component of the second pickup portion 2b, the second transfer platform portion 3b, and the second joining portion 4b and the symbols of each element of the first pickup portion 2a, the first transfer platform portion 3a, and the first joining portion 4a The symbol "a" corresponds.

First, the die supply unit 1 supplies the die D of the substrate P which is an example of assembly to the workpiece. The die supply unit 1 includes a wafer holding table 12 for holding the divided wafer 11, a lift unit 13 for lifting the die D from the wafer 11, and a wafer ring supply unit 18. The die supply unit 1 is moved in the XY direction by a driving means not shown, and the picked-up die D is moved to the position of the lifting unit 13. The wafer ring supply unit 18 has a wafer cassette (not shown) in which the wafer ring 14 is accommodated, and the wafer ring 14 is sequentially supplied to the die supply unit 1 and replaced with a new wafer ring. The wafer cassette accommodating the wafer ring 14 is supplied from the outside of the flip chip bonder 10 to the wafer ring supply unit 18. The die supply unit 1 moves the wafer ring 14 to the pickup point so that the desired die D can be picked up from the wafer ring 14. The wafer ring 14 is for fixing the wafer 11 and can be assembled to a jig of the die supply unit 1.

The first pick-up part 2a is located on the side of the substrate supply part 6K with respect to the die D to be picked up. The first pick-up part 2a includes a pick-up and turn head 21a that picks up the die D and reverses it, and a drive part 23a that lifts, rotates, reverses, and moves the collet 22a (see FIG. 11) in the X axis direction. And a pickup head 25a having a suction nozzle 26a (refer to FIG. 11) that sucks and holds the die D at the tip, and a drive part 27a that moves the pickup head 25a up and down and moves in the X-axis direction. A wafer recognition camera 24 is provided directly above the picked-up die D for the first pickup part 2a and the second pickup part 2b to share. With this configuration, as shown in FIG. 11, the pick-up and turn head 21a picks up the die D based on the imaging data of the wafer recognition camera 24, and the pick-up and turn head 21a is rotated 180 degrees to reverse the bump of the die D While facing downwards, the die D is set to the posture of being delivered to the pickup head 25a. The pickup head 25a receives the inverted die D from the pickup and inversion head 21a, and is placed on the first transfer platform part 3a.

The first transfer platform portion 3a includes transfer platforms 31a1, 31a2 on which the die D is temporarily placed, a bottom view camera (first imaging device) 34a, and a bottom view correction mark 35a. The transmission platforms 31a1, 31a2 can be moved in the Y-axis direction by a driving part not shown.

The first bonding portion 4a is located on the side of the substrate supply portion 6K with respect to the picked-up die D. The first bonding portion 4a picks up the die D from the transfer platforms 31a1 and 31a2, and bonds it to the substrate P that has been transported. The first joint 4a includes a joint head 41a, a joint head table 45a that moves the joint head 41a in the Z-axis direction, and a gantry table (Y Beam) 43a, a pair of X beams (not shown) that move the door-shaped table 43a in the X-axis direction, and a bonding camera (second) that photographs the position recognition mark (not shown) of the substrate P to recognize the bonding position Camera) 44a. The door-shaped table 43a extends in the Y-axis direction so as to straddle the assembly platform BS (refer to FIG. 12), and its two ends are supported by a pair of X beams freely in the X-axis direction. When the die D is bonded to the substrate P, the substrate P is sucked and fixed to the assembly platform BS. The bonding camera 44a is provided on the bonding head table 45a. As shown in FIG. 12, the bonding head 41a has bonding heads 41a1, 41a2, 41a3, 41a4, and each has suction nozzles 42a1, 42a2, 42a3, 42a4, and suction-holds the four die|dye D at the tip. With this configuration, the bonding head 41a picks up the die D from the transfer platforms 31a1, 31a2, and the bottom-view camera 34a and the bonding camera 44a photograph the position of the bonding head holding the die D. Based on the photographed data, the bonding positioning correction position is calculated, and the bonding head is moved to bond the die D to the substrate P.

The transport unit 5 is provided with transport rails 51 and 52 for moving the substrate P in the X-axis direction. The conveyance rails 51 and 52 are arranged in parallel. With such a configuration, the substrate P is carried out from the substrate supply part 6K, moved to the bonding position along the conveyance rails 51 and 52, moved to the substrate unloading section 6H after bonding, and delivered to the substrate unloading section 6H. In bonding the die D to the substrate P, the substrate supply unit 6K unloads the new substrate P and stands by on the conveyance rails 51 and 52. The substrate P is carried into the substrate supply section 6K from the outside of the flip chip bonder 10, and the substrate P on which the die D is placed is carried out from the substrate unloading section 6H to the outside of the flip chip bonder 10.

The control device 7 is provided with a memory device (memory) for storing programs (software) or data for monitoring and controlling the actions of various parts of the flip chip bonder 10, and a central processing unit (CPU) for executing the programs stored in the memory. The control device 7 includes the controller 101 of the embodiment, the recognition processing device 105, and the like.

As shown in FIG. 13, the wafer holding table 12 has an expansion ring 15 for holding the wafer ring 14, and a dicing tape 16 with a plurality of die D adhered to the wafer ring 14 for horizontal positioning support The ring 17 and the lifting unit 13 for lifting the die D upward. In order to pick up the predetermined die D, the lifting unit 13 is moved in the vertical direction by a driving mechanism not shown, and the die supply unit 1 is moved in the horizontal direction.

The configuration of the bottom-view camera and the joint camera will be described using FIG. 14. Fig. 14(a) is a schematic diagram of the structure of a bottom view camera, and Fig. 14(b) is a schematic diagram of the structure of a bonded camera.

The bottom view camera 34a includes an imaging device 344 having a camera body 344a and a lens 344b, and an illuminating device 343 having a pseudo-coaxial illumination 343a and a ring-shaped illumination 343b. The bottom-view camera 34a is arranged facing upward so as to photograph the upper die D from directly below. The camera body 344a and the lighting device 343 are controlled like the camera body 104a and the lighting device 103 of the imaging system 100 of the embodiment. That is, the bottom view camera 34a can shoot the die D with the single exposure of the strobe light or the multiple exposure of the strobe light according to the embodiment according to the instruction of the control device 7. The bottom view camera 34b and the bottom view camera 34a have the same configuration.

The bonded camera 44a includes an imaging device 444 having a camera body 444a and a lens 444b, and an illuminating device 443 having pseudo-coaxial illumination 443a and side illumination 443b. The bonding camera 44a is arranged face down so as to photograph the bottom view correction mark 35a, the die D, and the substrate P from above. The camera body 444a and the lighting device 443 are controlled like the camera 104a and the lighting device 103 of the imaging system 100 of the embodiment. That is, the bonding camera 44a can photograph the bottom-view correction mark 35a, the die D, and the substrate P through the single exposure of the strobe light or the multiple exposure of the strobe light of the embodiment. The bonded camera 44b and the bonded camera 44a have the same configuration.

In addition, the wafer recognition camera 24 has the same structure as the bonding camera 44a, except that the wafer 11 (die D) is photographed by normal exposure, but it can also be replaced like the bottom view camera 34a or the bonding camera 44a. Strobe light exposure or strobe light multiple exposure system.

Use Fig. 15 and Fig. 16 to explain the general exposure shooting system using constant lighting and the camera shutter for exposure control. Fig. 15 is a schematic diagram of the composition of a general exposure shooting system. FIG. 16 is a schematic diagram of the operation time point of the photographing system of FIG. 15.

The imaging system 100R for general exposure includes a controller 101, lighting power supplies 102a, 102b, an illuminating device 103, an imaging device 104 having a camera 104a and a lens 104b, and a recognition processing device 105.

The controller 101 outputs control signals such as lighting ON/OFF and illuminance setting to the lighting power supplies 102a and 102b, and the lighting power supplies 102a and 102b supply the lighting current to the lighting device 103. The lighting device 103 is composed of pseudo-coaxial lighting 103a and side lighting (ring-shaped lighting, 4-directional rod-shaped lighting, etc.) 103b, and the like. The illuminating device 103 is to arrange a plurality of light-emitting diodes neatly in a matrix, and to illuminate the workpiece 108 of the photographing object by the light source of the surface light.

The lighting device 103 is always turned on or turned on before the recognition exposure. The controller 101 inputs a shooting trigger signal to the camera 104a. After the exposure delay time (Td1), the camera 104a opens the shutter to expose the set exposure time (Tex3), and shoots the workpiece 108 of the subject through the lens 104b. After the photographing of the camera 104a is completed (after the exposure is finished, the transfer waiting time (Td2)), the camera 104a transfers the image data to the recognition processing device 105, and the recognition processing device 105 calculates the positional deviation amount based on the image data.

The recognition time (Tr4) is the same as in the first embodiment and is as follows. Tr4=Td1＋Tex3＋Td2＋Ttr＋Td3＋Trc However, the exposure time (Tex3) is longer than the exposure time (Tex1) of the first embodiment, so the recognition time (Tr3) will be exactly longer than the recognition time (Tr) of the first embodiment (Tex3-Tex1).

As mentioned above, in a device where multiple axes are operating at the same time, when recognizing the movement of other axes during shooting, due to the influence of motion vibration, the captured image may be in a static position or when it is vibrating. There will be positional deviation. However, as shown in Figure 16, by extending the exposure time, the shaking caused by the vibration in the captured image will be averaged, and the influence of the vibration can be reduced. In addition, by ensuring that the exposure time is above 10ms, the motion vibration above 100Hz will be averaged by shooting, and the impact of vibration can be reduced.

Next, the bottom view correction will be explained using FIGS. 17 and 18. Fig. 17 is an explanatory diagram of the operation of the bonding head when viewed from the direction of arrow B in Fig. 10, Fig. 17(a) is a schematic diagram of the state of photographing the first bonding head and its corresponding bottom-view correction mark, Fig. 17(b ) Is a schematic diagram of the state of shooting the second bonding head and its corresponding bottom-view correction mark, Figure 17(c) is a schematic diagram of the state of shooting the third bonding head and its corresponding bottom-view correction mark, Figure 17 (d) is a schematic diagram showing the state of shooting the fourth joint head and its corresponding bottom-view correction mark. Figure 18 (a) is a schematic diagram of the captured image of the bottom view correction mark made by the bonded camera, Figure 18 (b) is a schematic diagram of the captured image of the die made by the bottom view camera, and Figure 18 (c) is the schematic diagram of the captured image of the bottom view camera. The taken image of the reference mark on the made substrate. In addition, in the captured image of FIG. 18, the outer periphery is shown in black and the inner is shown in white for convenience of explanation.

The bottom view correction mark 35a is provided on the side opposite to the substrate P with the bottom view camera 34a interposed therebetween. As shown in Fig. 17, the bottom view correction mark 35a is formed as a mark made by a point light source, that is, a pinhole of the light source 35a6 to 35a9 such as an LED arranged under the plate 35a5 that shields light against the plate 35a5. 35a1 to 35a4 are irradiated. Four pinholes 35a1 to 35a4 are arranged at positions corresponding to the four suction nozzles 42a1 to 42a4. Therefore, the four pinholes 35a1 to 35a4 constitute four bottom view correction marks. In addition, when referring to each of the four bottom view correction marks, symbols 35a1 to 35a4 are used.

When the bottom-view correction mark 35a is photographed by the joint camera 44a, the lighting device 443 of the joint camera 44a will be turned off to prevent the light emitted from the lighting device 443 of the joint camera 44a from being incident on the bottom-view camera 34a to recognize the bottom-view camera 34a Make an impact. Thereby, it is possible to reduce the influence of the photographing of the suction nozzle 42a by the bottom-view camera 34a described later.

The bonding head 41a and the bonding camera 44a move in the Y-axis direction, and the bonding camera 44a photographs and recognizes the bottom view correction mark 35a. As shown in FIG. 18(a), the position of the initially registered bottom view correction mark 35a is recognized (mark initial The deviation between the position) and the position of the bottom view correction mark 35a recognized by the bonding camera 44a is calculated (dXmark, dYmark). When the suction nozzle 42a is positioned at the center of the bottom view camera 34a, the bottom view correction mark 35a is photographed by the joint camera, and the position is measured to register the initial position of the mark in advance. At the same time that the bottom-view correction mark 35a is photographed by the camera 44a, the bottom-view camera 34a photographs and recognizes the mounting surface of the die D sucked and held by the nozzle 42a, as shown in FIG. 18(b), from the bottom The position of the measured die D is recognized and calculated by the center of the camera 34a (dXdie, dYdie). From the results of (dXmark, dYmark) and (dXdie, dYdie), the position of the die D adsorbed and held by the suction nozzle 42a relative to the position of the bottom view correction mark 35a is calculated.

As shown in FIG. 17, for the four suction nozzles 42a1 to 42a4 mounted on the bonding head 41a, the die D held by each is moved to the bottom view camera 34a, and the die D is photographed to identify the position. 42a4, while recognizing the shooting die D by the bottom view camera 34a, the bottom view correction marks 35a1 to 35a4 corresponding to the positions recognized by the bottom view camera 34a are photographed and recognized by the bonding camera 44a. The following is a detailed description.

First, as shown in Figure 17(a), the die D sucked and held by the nozzle 42a1 of the first bonding head 41a1 is moved to the top of the bottom-view camera 34a, the die D is photographed to identify the position, and the die The die D held by the suction nozzle 42a1 is sucked, and the position is recognized by the bottom view correction mark 35a1 corresponding to the position captured by the camera 44a and recognized by the bottom view camera 34a.

Next, as shown in FIG. 17(b), the die D sucked and held by the suction nozzle 42a2 of the second bonding head 41a2 is moved to the top of the bottom view camera 34a, the die D is photographed to identify the position, and the The die D held by the suction nozzle 42a2 is sucked, and the position is recognized by the bottom view correction mark 35a2 corresponding to the position captured by the camera 44a and recognized by the bottom view camera 34a.

Next, as shown in FIG. 17(c), the die D sucked and held by the suction nozzle 42a3 of the third bonding head 41a3 is moved to the top of the bottom-view camera 34a, the die D is photographed to identify the position, and the The die D held by the suction nozzle 42a3 is sucked, and the position is recognized by the bottom view correction mark 35a3 corresponding to the position captured by the camera 44a and recognized by the bottom view camera 34a.

Next, as shown in FIG. 17(d), the die D sucked and held by the suction nozzle 42a4 of the fourth bonding head 41a4 is moved to the top of the bottom-view camera 34a, the die D is photographed to identify the position, and the die D The die D held by the suction nozzle 42a4 is sucked and the position is recognized by the bottom view correction mark 35a4 corresponding to the position captured by the camera 44a and recognized by the bottom view camera 34a.

After the position identification of the die D by the bottom-view camera 34a is completed, as shown in FIG. 17, the bonding head 41a moves to the substrate P which is adsorbed and fixed to the assembly platform BS while holding the die D. The die D is joined at the predetermined position. At this time, the reference mark RM provided on the substrate P is recognized by the bonding camera 44a, and the position of the reference mark RM on the substrate P when viewed from the center of the bonding camera 44a is recognized and measured as shown in FIG. 18(c), and Calculate (dX _RM , dY _RM ). When calculating the position of the bonding die D from the position of the reference mark RM, the position of the die D sucked and held by the suction nozzle 42a relative to the position of the bottom view correction mark 35a obtained previously is added to calculate the bonding position. That is, each recognition result is subtracted from the target positioning coordinates to be joined to calculate (Xbond, Ybond).

Xbond = dX _RM- dXdie-dXmark Ybond = dY _RM- dYdie-dYmark. By this, it can offset the influence of the positioning error of the bonding head, or the deviation of the bonding head positioning position caused by the action vibration, and always look at the bottom The correction mark 35a is used as a reference to calculate the position of the die D, and the bonding position can be matched from the reference mark RM. In this example, when the head is positioned at the position to be joined by the nozzles, the reference mark RM can be set in the field of view that can be photographed by the joining camera and recognized to correct the joining position. The method of correcting the positioning position during bonding is to bond the camera by setting the alignment mark in advance on the substrate to be bonded, or individually setting marks on each bonding position on the substrate, or using the bonded die as a reference. The same effect can be obtained by performing shooting recognition calculations and matching the recognition results of the bottom view correction marks to correct the joint position.

Next, the bonding method (the manufacturing method of the semiconductor device) implemented in the flip chip bonder of the embodiment will be described using FIG. 19. 19 is a schematic flow chart of the bonding method implemented in the flip chip bonder of the embodiment. The description is mainly based on the side of the first pickup portion 2a, the first transfer platform portion 3a, and the first joining portion 4a, but the second pickup portion 2b, the second transfer platform portion 3b, and the second joining portion 4b side are also the same. In addition, the second pickup portion 2b, the second transfer platform portion 3b, and the second joint portion 4b operate in parallel within a range that does not interfere with the first pickup portion 2a, the first transfer platform portion 3a, and the first joint portion 4a.

In the case of manufacturing a semiconductor device by the flip chip bonder 10, the wafer cassette containing the wafer ring 14 is supplied to the wafer ring supply part 18 from the outside of the flip chip bonder 10. In addition, the substrate P is carried into the substrate supply portion 6K from the outside of the flip chip bonder 10. The substrate P on which the die D is assembled is carried out from the substrate carrying-out portion 6H to the outside of the flip chip bonder 10. The operation in the flip chip bonder 10 will be described below.

Step S1: The control device 7 moves the wafer holding table 12 so that the die D to be picked up is located directly above the jacking unit 13, and based on the imaging data of the wafer recognition camera 24, the peeling target die is positioned on the jacking unit 13 and Suction nozzle 22a. The lifting unit 13 is moved so that the upper surface of the lifting unit 13 contacts the back surface of the dicing tape 16. At this time, the control device 7 sucks the dicing tape 16 onto the upper surface of the lifting unit 13. The control device 7 evacuates the suction nozzle 22a while lowering it, landing on the crystal grain D to be peeled off, and adsorbing the crystal grain D. The control device 7 raises the suction nozzle 22a to peel the die D from the dicing tape 16. Thereby, the die D is picked up by the picking and turning head 21a.

Step S2: The control device 7 moves the pick-up and turn head 21a.

Step S3: The control device 7 rotates the pick-up and inverting head 21a by 180 degrees, inverts the bump surface (surface) of the die D and faces downward, and inverts the bump (surface) of the die D and faces downward, to complete The die D is delivered to the posture of the pickup head 25a.

Step S4: The control device 7 picks up the die D from the suction nozzle 22a of the pick-up and turn head 21a through the suction nozzle 26a of the pickup head 25a, and transfers the die D.

Step S5: The control device 7 reverses the pick-up and inversion head 21a, and turns the suction surface of the suction nozzle 22a downward.

Step S6: Before or in parallel with step S5, the control device 7 moves the pickup head 25a to the transfer platform 31a1.

Step S7: The control device 7 places the die D held by the pickup head 25a on the transfer platform 31a1.

Step S8: The control device 7 moves the pickup head 25a to the receiving position of the die D. The control device 7 repeats steps S1 to S8 a predetermined number of times (4 times in the embodiment).

Step S9: After step S8 or in parallel, the control device 7 moves the transfer platform 31a1 to the receiving position with the bonding head 41a.

Step SA: The control device 7 photographs the die D held on the transfer platform 31a1 by the bonding camera 44a, and collectively picks up plural (four in the embodiment) die from the transfer platform 31a1 by the suction nozzle 42a of the bonding head 41a D, accept and accept the die D.

Step SB: The control device 7 moves the four die D held by the suction nozzles 42a1, 42a2, 42a3, 42a4 of the bonding heads 41a1, 41a2, 41a3, 41a4 from the transfer platform 31a1 to the substrate P. At this time, the bottom-view correction mark 35a is photographed while the bonding camera 44a is moved, and the four moving die D is photographed by the bottom-view camera 34a. The door-shaped table 43a moves in the X-axis direction and the bonding head 41a moves in the Y-axis direction.

Step SC: The control device 7 photographs the substrate by the bonding camera 44a, based on the data of the bottom view correction mark 35a captured by the bonding camera 44a, the data of the die D captured by the bottom view camera 34a, and by bonding the camera The material of the substrate photographed by 44a is collectively placed on the substrate P with four die D picked up from the transfer platform 31a1 by the suction nozzles 42a1, 42a2, 42a3, and 42a4 of the bonding heads 41a1, 41a2, 41a3, and 41a4.

Step SD: The control device 7 moves the bonding heads 41a1, 41a2, 41a3, 41a4 to the receiving and receiving position with the transfer platform 31a2.

Step SE: The control device 7 moves the transfer platform 31a1 to the receiving position with the pickup head 25a.

In addition, in the above process, although it is an example of using the delivery platform 31a1, the same is true in the case of using the delivery platform 31a2.

In the flip-chip bonder 10, in the shooting of the die and the substrate, by using a light-emitting strobe light of about 10 μs, the exposure shooting is performed during the movement of the shooting object, that is, the die and the substrate (flight scan It can improve the productivity, but if it is a very short time strobe light exposure, it will have the influence of the identification deviation caused by the influence of external disturbance such as vibration, which may bring about the high-precision joining Hinder. In such a situation, the exemplified strobe light multiple exposure proposed by the present invention is to use the camera to expose for more than 10ms, and the strobe light will be illuminated multiple times during the exposure time to reduce the identification caused by the impact of vibration during shooting. It is possible to prevent errors in the identification correction from deteriorating accuracy and provide high-precision joining operations.

Above, the invention completed by the present invention team has been specifically explained based on the embodiments and examples, but the present invention is of course not limited to the above-mentioned embodiments and examples, and various changes can be made.

In the embodiment, an example in which the controller 101 outputs the shooting trigger signal to the camera 104a is described, but it can also be designed such that the recognition processing device 105 outputs an exposure start instruction to the camera 104a, and the camera 104a that receives the exposure start instruction starts exposure and then 104a outputs the lighting trigger signal to the strobe lighting power supplies 107a, 107b.

In addition, the embodiment and the examples illustrate the example of the illuminating device composed of pseudo-coaxial lighting and ring-type lighting, but it can also be composed of one or the other of pseudo-coaxial lighting and ring-type lighting, and combining them. Constituted by lighting.

In addition, in the embodiment, an example in which there are four bonding heads (assembly heads) is described, but it is not limited to this, and it may be one or more bonding heads.

In addition, the embodiment describes an example in which the reversing mechanism is set on the pick-up and turn head, and the transfer head receives the die from the pick-up and turn head and places it on the intermediate platform, and moves the intermediate platform, but it is not limited to this, and can be designed as The pick-up and turn head that moves and picks up the die and reverses can also be designed to place the picked-up die D on a platform unit that can rotate the front and back of the die to move the platform unit.

In addition, the embodiments illustrate examples of applying flip chip bonders, but they are not limited to this. They can also be applied to die bonding machines that bond semiconductor chips (die) to substrates, or to assemble packaged semiconductor devices on the substrates. The chip laminator (surface mounter), etc.

100: shooting system 101: Controller 103: lighting device 104: Camera 105: identification processing device 107: Strobe lighting power supply 108: Workpiece (subject)

[Fig. 1] Fig. 1 is a schematic diagram of the configuration of the imaging system of the first embodiment. [FIG. 2] FIG. 2 is a schematic diagram of the action time point of the shooting system in FIG. [Fig. 3] Fig. 3 is a schematic diagram of light-emitting time points of the photographing system of Fig. 1. [Fig. [Fig. 4] Fig. 4 is a schematic diagram of the action time point in the case of shooting from the stop of the axis movement of the photographing system of Fig. 1 until a recognition waiting time is established before the recognition starts. [FIG. 5] FIG. 5 is a schematic diagram of the action time point of the shooting system of FIG. [Fig. 6] Fig. 6 is a schematic diagram of the configuration of the imaging system of the second embodiment. [Fig. 7] Fig. 7 is a schematic diagram of the action time point of the photographing system of Fig. 6. [Fig. [Fig. 8] Fig. 8 is a schematic diagram of the light-emitting time point of the photographing system of Fig. 6. [Fig. 9] Fig. 9 is a schematic diagram of the configuration of the imaging system of the first modification. [Fig. 10] Fig. 10 is a schematic plan view of the flip chip bonder of the embodiment. [Fig. 11] Fig. 11 is an explanatory diagram of the operation of the pick-up and reversal head and the transfer head when viewed from the direction of arrow A in Fig. 10; [Fig. 12] Fig. 12 is an explanatory diagram of the operation of the bonding head when viewed from the direction of arrow B in Fig. 10. [Fig. 13] Fig. 13 is a schematic schematic cross-sectional view of a main part of the crystal grain supply part of Fig. 10. [Fig. 14] Fig. 14(a) is a schematic diagram of the structure of a bottom view camera, and Fig. 14(b) is a schematic diagram of the structure of a bonded camera. [Fig. 15] Fig. 15 is a schematic diagram of the structure of a shooting system for general exposure. [FIG. 16] FIG. 16 is a schematic diagram of the operation time point of the photographing system of FIG. 15. [Fig. 17] Fig. 17 is an explanatory diagram of the operation of the bonding head when viewed from the direction of arrow B in Fig. 10. [Fig. 18] Fig. 18 is a schematic diagram of an image of a joined camera and a bottom view camera. [FIG. 19] FIG. 19 is a schematic flowchart of the bonding method implemented in the flip chip bonder of the embodiment.

Claims (17)

An assembly device with: The first photographing device for the first photographic object to move relatively to photograph the aforementioned first photographic object; and A first lighting device that irradiates light to the aforementioned first photographic object; The first illuminating device is configured to emit a plurality of strobe lights in a period less than 1/2 of the vibration period during the exposure time of the first imaging device. Such as the assembly device of claim 1, wherein It is also equipped with a stroboscopic lamp lighting power supply for supplying lighting luminescence current to the first lighting device, The first imaging device is configured to start exposure based on a shooting trigger signal, and to generate a strobe light trigger signal that is supplied to the strobe light illumination power source, The strobe lighting power supply is configured to periodically supply the lighting current to the first lighting device based on the strobe trigger signal. Such as the assembly device of claim 1, which further has: Strobe lighting power supply for supplying lighting luminous current to the aforementioned first lighting device; and Based on the trigger signal, a pulse wave generating circuit that generates the strobe light trigger signal supplied to the aforementioned strobe light illumination power source; The first imaging device is configured to start exposure based on a shooting trigger signal, and to generate a strobe light trigger signal that is supplied to the strobe light illumination power source, The strobe lighting power supply is configured to periodically supply the lighting current to the first lighting device based on the strobe trigger signal. Such as the assembly device of claim 1, which further has: The second photographing device that moves the second photographic object relatively to photograph the aforementioned second photographic object; and A second lighting device that irradiates the aforementioned second photographic object with light; The second lighting device is configured to emit a plurality of strobe lights at a period of 1/2 or less of the vibration period during the exposure time of the second imaging device. Such as the assembly device of claim 4, wherein, The aforementioned first photographic object is a die, It also has a first assembly head for transporting the aforementioned die, The first imaging device is configured to be fixed, and to photograph the die of the moving first assembly head from below. Such as the assembly device of claim 5, wherein, The aforementioned second photographic object is a correction mark, The second imaging device is configured to be mounted on the first assembly head, and when the first imaging device captures the die held by the moving first assembly head, the correction mark is captured from above. Such as the assembly device of claim 6, wherein The second imaging device is configured to image the substrate on which the die is placed from above. Such as the assembly device of claim 7, wherein It also has a transfer platform to transport the aforementioned dies, The second imaging device is configured to image the crystal grains held on the transfer platform from above. Such as the assembly device of claim 4, wherein, The aforementioned first lighting device has pseudo-coaxial lighting and side lighting, The aforementioned second lighting device includes pseudo-coaxial lighting and side lighting. Such as the assembly device of claim 8, which further has: Assembly platform; and A first beam that extends in the first direction so as to straddle the above-mentioned assembly platform, and the two ends of the first beam are supported so as to move freely in the second direction; and A second beam that extends in the first direction so as to straddle above the assembly platform, and whose two ends are each freely movable in the second direction and are supported above the assembly platform; The first assembly head is configured to be supported by the first beam movably in the first direction. Such as the assembly device of claim 10, which further has: An inverted pickup head that picks up the aforementioned die from the die supply part and reverses it; and A pickup head capable of moving freely in the first direction to pick up the die picked up by the flip pickup head; The transfer platform can freely move in the second direction, and the die picked up by the pickup head is placed, The first assembly head is configured to pick up the die placed on the transfer platform and place the substrate on the assembly platform. Such as the assembly device of claim 11, wherein The first imaging device is configured to image the die from below when the first assembling head transports the die from the transfer platform to the assembly platform. Such as the assembly device of claim 11, wherein The second imaging device is configured to capture the correction mark from above when the first assembly head transports the die from the transfer platform to the assembly platform. Such as the assembly device of claim 11, wherein It is further provided with a second assembling head supported by the second beam movably in the first direction, The second assembly head is configured to pick up the die placed on a transfer platform different from the transfer platform and place the substrate on the assembly platform. A method of manufacturing a semiconductor device, including: The process of transporting the aforementioned substrate into an assembly device, wherein the assembly device is provided with a first imaging device for imaging the die, a second imaging device for imaging the substrate, a first illuminating device for irradiating light to the die, and irradiating the substrate The second lighting device for light and the assembly head for transporting the aforementioned die; and When the die is being transported by the assembly head, the first photographing process in which the first imaging device photographs the die from below; and The second photographing process in which the aforementioned second photographing device photographs the aforementioned substrate from above; and The process of placing the aforementioned die on the aforementioned substrate based on the data taken in the aforementioned first shooting process and the aforementioned second shooting process; In the aforementioned first photographing project, the aforementioned first illuminating device emits a plurality of strobe lights with a period of less than 1/2 of the vibration period during the exposure time of the aforementioned first photographing device, and the aforementioned crystal particles are photographed, In the aforementioned second photographing process, the aforementioned second illuminating device emits a plurality of strobe lights at a period of less than 1/2 of the vibration period within the exposure time of the aforementioned second photographing device to photograph the aforementioned substrate. The method of manufacturing a semiconductor device according to claim 15, wherein In the first photographing process, the first photographing device photographs the crystal grain, and the second photographing device photographs the bottom view correction mark, and the positioning position of the crystal grain relative to the first photographing device is identified and corrected. The method for manufacturing a semiconductor device according to claim 15, which further includes: Move into the wafer ring project; and The third photographing process of photographing the dies in the aforementioned wafer ring; In the aforementioned third photographing process, the aforementioned crystal grains are photographed by performing constant illumination during the exposure time.

Images