An adaptive error correction method in wafer picking process

Abstract

The present invention discloses an adaptive error correction method in a wafer fetching process, which utilizes a camera module to realize the acquisition operation of parameter data in the wafer fetching process, and utilizes a wafer suction and mounting module to realize the wafer fetching operation, and the present invention relates to the field of wafer mounting technology. The adaptive error correction method in the wafer fetching process receives the collected data and performs preliminary processing, establishes an error compensation model through equipment parameters, and then introduces the processed data into the error compensation model for analysis operation, combines the operation error of the mounting module and the positioning error of the camera module affected by temperature, and obtains the adjustment result data of the mounting module braking and the camera welding head center reference, thereby realizing the adaptive error correction adjustment operation in each wafer fetching process, ensuring the integrity and accuracy of the wafer in the wafer fetching process, so that the wafer fetching operation can be completed in a shorter time, and improving the overall efficiency of the wafer processing production line.

Description

Technical Field

[0001] The invention relates to the technical field of wafer mounting, and in particular to an adaptive error correction method in a wafer picking process. Background Art

[0002] In the wafer processing of semiconductor manufacturing, the wafer picking operation is crucial. However, due to the influence of various factors, errors will inevitably occur in the wafer picking process.

[0003] The reference patent name is: A vision-based automatic correction method for wafer angle deviation (patent publication number: CN103107121A, patent publication date: 2013-05-15). The wafer is set on the suction plate of the four-axis CNC machine platform system, and the image of the wafer taken by the microscope camera of the four-axis CNC machine platform system is detected and positioned using Hough transform. An image is taken at point A of the wafer to determine the position of the characteristic straight line and characteristic point on the wafer. According to the angle formed by the characteristic straight line and the X or Y axis direction of the four-axis CNC machine platform, the position of the wafer is adjusted. Images are taken again at points B and C of the wafer for Hough change detection. By judging whether the inclination angle β of the line connecting the midpoints of the characteristic straight line at points B and C is 0, it is verified whether the wafer is accurately positioned. The wafer position and cutting path can be accurately positioned, with high reliability and fast response speed, which promotes the automation and intelligence of wafer dicing technology. The effective accuracy is controlled at about 2μ, shortening the positioning time of the wafer and improving the processing efficiency of the wafer.

[0004] Based on the description in the above-mentioned document, in the existing wafer picking process, in theory, fixed-point repeated placement operations can be achieved each time by matching the chip to be picked up with the cross center cursor of the camera. However, in reality, the position of the cross center cursor of the camera changes due to the influence of temperature and stress, so that when performing the placement operation, the center position of the welding head does not correspond to the chip position. If the operation is performed according to the theory, there will be a deviation problem in the placement position. For this reason, the present invention provides an adaptive error correction method in the wafer picking process. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides an adaptive error correction method in a wafer picking process, which solves the problem that in the prior wafer picking process, in theory, fixed-point repeated placement operations can be achieved each time by matching the chip to be picked up with the cross center cursor of the camera, but in reality, the position of the cross center cursor of the camera changes due to the influence of temperature and stress, so that when performing the placement operation, the center position of the welding head does not correspond to the chip position, and if the theoretical operation is followed, a deviation problem of the placement position will occur.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an adaptive error correction method in a wafer fetching process, specifically comprising the following steps:

[0007] A1. Use the camera module to collect parameter data during wafer pickup, and use the wafer suction and placement module to pick up the wafer and transmit the operating parameters, while the camera module and the placement module keep moving synchronously.

[0008] A2. Receive the collected data and perform preliminary processing, establish an error compensation model through equipment parameters, and then introduce the processed data into the error compensation model for analysis and operation. Combined with the operation error of the placement module and the positioning error of the camera module affected by temperature, the adjustment result data of the placement module braking and the camera welding head center reference are obtained;

[0009] A3. Generate correction instructions based on the result data and transmit them to the camera module and the suction and placement module to implement the adjustment operation. By adjusting the working temperature, the errors at different temperatures are repeatedly calculated to achieve predictive adjustment during application, and the data generated by the operation is displayed on the display panel.

[0010] Preferably, the operation of receiving the collected data and performing preliminary processing in A2 is:

[0011] a21. After receiving the collected data, classify the data according to subsequent needs and perform preliminary processing of the data;

[0012] a22. The data required subsequently are the operation parameter data of the suction and placement module and the image data collected by the camera module, and the redundant data is removed;

[0013] a23. Summarize the data to form operating parameter data sets and image data sets.

[0014] Preferably, the steps of establishing the error compensation model are:

[0015] B1. First, scale the camera module parameters, the chip placement module parameters, and the size and position parameters of the wafer and chip to form a three-dimensional virtual operation model;

[0016] B2. Then, the operating principle of the camera module and the suction and placement module is combined with the three-dimensional virtual operation model to realize dynamic operation, and then the real-time temperature data is introduced into the three-dimensional virtual operation model to become an error compensation model;

[0017] B3. The processed operating parameter data set and image data set are transmitted to the error compensation model for analysis.

[0018] Preferably, the steps of performing the analysis operation of the error compensation model in A2 are:

[0019] C1. Introduce the operating parameter data set and the image data set into the error compensation model for analysis;

[0020] C2, realize the return operation after the first wafer pick-up and placement, and analyze the camera captured images during the return positioning;

[0021] C3, by recording the running track of the suction and placement module, and calculating the running error generated;

[0022] C4. Compensation instructions are formed based on the numerical results of the error calculation to implement correction operations on the camera module and the pickup and placement module.

[0023] Preferably, the analysis operation of the collected image of the return positioning in C2 is:

[0024] c21, graying the collected image data, and then dividing the characteristic parts of the image data according to the gray value;

[0025] c22. First extract the welding head features after the previous chip is welded and sent out, and then extract the chip features after the next chip is sucked into the welding position, compare the center coordinates of the welding head features and the chip features, and determine the total error value.

[0026] Preferably, the coordinate comparison operation of the chip feature and the welding head feature in c22 is:

[0027] D1. Take the center position of the welding head feature as the coordinate center, establish the X-axis in the left and right directions of the welding head feature, and establish the Y-axis in the front and rear directions of the X-axis of the welding head feature;

[0028] D2. Draw cross lines at the four corner points of the chip feature, and the intersection of the lines is the center point of the chip feature. The center point is located in the XY coordinate system in D1 and is marked as Q. The coordinate is (x 1 ,y 1 ), and the center point Q is located in the third quadrant of the XY coordinate system;

[0029] D3, and the deviation distance between the center point Q and the characteristic center of the welding head is |x 1 |, the front and rear deviation distance of the center point Q from the welding head feature center is |y 1 |, while the left and right deviation distances and the front and rear deviation distances include the deviation of the welding head feature center position affected by external factors and the inertia deviation during operation.

[0030] Preferably, the calculation operation of the running error in C3 is:

[0031] c31, determine the running track of the suction and placement module, that is, the left and right side movement operation and the front and back side movement operation of the suction and placement module when performing the return operation;

[0032] c32. Calculate the inertia deviation in the left and right directions and the inertia deviation in the front and back directions through the operating parameters of the suction and placement module.

[0033] Preferably, the inertia deviation calculation operation in c32 is:

[0034] The calculation formula for the inertia deviation in the left and right directions is:

[0035] S 1 =(v 1 ×t 1 ) / 2;

[0036] And S 1 is the distance value of the inertial deviation in the left and right directions, v 1 is the inertia speed after braking to a specified distance, t 1 The time from the start of braking for the left and right side movement of the suction and placement module to the stop of the movement of the suction and placement module;

[0037] v 1 =P 1 / (F 牵 -F 制 );

[0038] P 1 In order to achieve the uniform movement of the suction and placement module, the driving power, F 牵 F is the traction force of the suction mount module during normal movement. 制 The braking force is used to stop the suction piece placement module, and the braking force is opposite to the traction force;

[0039] F 牵 =P 1 / v 2 ;

[0040] v 2 The speed at which the suction and placement module is pulled;

[0041] The calculation formula for the inertia deviation in the front and rear directions is:

[0042] S 2 =(v 3 ×t 3 ) / 2;

[0043] And S 2 is the distance value of the inertial deviation in the front and rear directions, v 3is the inertia speed after braking to a specified distance, t 3 It is the time from the start of braking operation for the front and rear side movement of the suction and placement module to the stop of the movement of the suction and placement module.

[0044] Preferably, the deviation of the characteristic center position of the welding head in D3 affected by external factors is:

[0045] The characteristic center position of the welding head is located at the left and right side deviation distance:

[0046] L 1 =S 1 +|x 1 |;

[0047] That is, adjust the left and right directions of the welding head feature center position, and compensate the deviation distance to the error compensation model to move L to the left. 1 distance;

[0048] The characteristic center position of the welding head is located in the front and rear direction. The deviation distance is:

[0049] L 2 =|y 1 |-S 2 ;

[0050] That is, the characteristic center position of the welding head is adjusted in the front and rear directions, and the deviation distance is compensated to move the error compensation model to the rear side by L 2 distance.

[0051] Preferably, the operation of calculating the error at different temperatures in A3 is:

[0052] E1. Determine the temperature range to be measured, the interval is [T min , T max ], and realize the collection of operating parameters. In this temperature range, select multiple temperature points marked as T i (i=1, 2, ..., n), and measure the placement error A of the placement machine at each temperature point i , and each temperature point T i And the corresponding error value A i Record it and form a data set {(T 1 , A 1 ), (T 2 , A 2 ),…,(T n , A n )};

[0053] E2. Establish a linear model between temperature change and error, and obtain the fitting function by solving it using the least squares method;

[0054] The relationship between the linear model error A and temperature T is:

[0055] A=β 0 +β 1 T;

[0056] Among them, β 0 and β 1 are the model parameters to be determined;

[0057] And construct the error sum of squares P:

[0058] ;

[0059] In order to find the β that minimizes P 0 and β 1 , find β for P 0 and β 1 The partial derivative of , and set the partial derivative equal to 0:

[0060] ;

[0061] ;

[0062] Arrange the two equations into a linear system and solve them to obtain β 0 and β 1 :

[0063] ;

[0064] E3. Use the obtained β 0 and β 1 The predicted error values ​​of the model at different temperatures are substituted into the relationship formula of the linear model, and compared with the actual measured values ​​to verify the operation of the model.

[0065] The present invention provides an adaptive error correction method in a wafer fetching process. Compared with the prior art, the method has the following beneficial effects:

[0066] 1. The adaptive error correction method in the wafer picking process receives the collected data and performs preliminary processing, establishes an error compensation model through equipment parameters, and then introduces the processed data into the error compensation model for analysis and operation. Combined with the operation error of the mounting module and the positioning error of the camera module affected by temperature, the adjustment result data of the mounting module braking and the camera welding head center reference are obtained, thereby realizing the adaptive error correction and adjustment operation in each wafer picking process, ensuring the integrity and accuracy of the wafer in the wafer picking process, enabling the wafer picking operation to be completed in a shorter time, and improving the overall efficiency of the wafer processing production line.

[0067] 2. The adaptive error correction method in the wafer picking process grayscales the collected image data, and then divides the feature parts in the image data according to the grayscale value, extracts the welding head features after the previous chip is welded and sent out, and then extracts the chip features after the next chip is absorbed to the welding position, compares the center coordinates of the welding head features and the chip features, and determines the total error value, so as to determine the error of the welding head features caused by various influences, thereby realizing the differentiated correction of the deviation of the center position of the welding head features affected by external factors and the inertial deviation in the running operation, realizing adjustment operations for different distances or situations, and ensuring the accuracy of the adjusted data.

[0068] 3. The adaptive error correction method in the wafer picking process determines the operation trajectory of the wafer suction and placement module, that is, the left and right side movement operation and the front and rear side movement operation of the wafer suction and placement module during the return operation, and calculates the inertia deviation in the left and right directions and the front and rear directions through the operation parameters of the wafer suction and placement module, thereby realizing early braking to eliminate the inertia deviation, and then adjusting the center position of the welding head of the wafer suction and placement module to avoid the influence of the inertia deviation on the adjustment accuracy of the center position of the welding head, so as to make the error adjustment more efficient and improve the adaptability and operation stability in the wafer picking process. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 is an operation flow chart of the adaptive error correction method of the present invention;

[0070] Figure 2 is a flow chart for establishing the error compensation model of the present invention;

[0071] Figure 3 An operational flow chart for error compensation model analysis of the present invention;

[0072] Figure 4 Schematic diagram of the position error of the welding head of the present invention. DETAILED DESCRIPTION

[0073] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] See also Figure 1-Figure 4 , the present invention provides two technical solutions:

[0075] Embodiment 1: A method for adaptive error correction in a wafer fetching process, comprising the following steps:

[0076] A1. Use the camera module to collect parameter data during wafer pickup, and use the wafer suction and placement module to pick up the wafer and transmit the operating parameters, while the camera module and the placement module keep moving synchronously.

[0077] A2. Receive the collected data and perform preliminary processing, establish an error compensation model through equipment parameters, and then introduce the processed data into the error compensation model for analysis and operation. Combined with the operation error of the placement module and the positioning error of the camera module affected by temperature, the adjustment result data of the placement module braking and the camera welding head center reference are obtained;

[0078] A3. Generate correction instructions based on the result data and transmit them to the camera module and the suction and placement module to implement the adjustment operation. By adjusting the working temperature, the errors at different temperatures are repeatedly calculated to achieve predictive adjustment during application, and the data generated by the operation is displayed on the display panel.

[0079] By receiving the collected data and performing preliminary processing, an error compensation model is established through equipment parameters, and then the processed data is introduced into the error compensation model for analysis. Combined with the operating error of the placement module and the positioning error of the camera module affected by temperature, the adjustment result data of the placement module braking and the camera welding head center reference are obtained, thereby realizing adaptive error correction and adjustment operations in each wafer picking process, ensuring the integrity and accuracy of the wafer during the wafer picking process, allowing the wafer picking operation to be completed in a shorter time, and improving the overall efficiency of the wafer processing production line.

[0080] In the embodiment of the present invention, the operation of receiving the collected data and performing preliminary processing in A2 is:

[0081] a21. After receiving the collected data, classify the data according to subsequent needs and perform preliminary processing of the data;

[0082] a22. The data required subsequently are the operation parameter data of the suction and placement module and the image data collected by the camera module, and the redundant data is removed;

[0083] a23. Summarize the data to form operating parameter data sets and image data sets.

[0084] Among them, the specific data classification is achieved by extracting the category subject of the subsequent required data, and then matching the category subject with the type title of the collected data. After matching, the same type title or similar type title will introduce the corresponding collected data into the result column of the category subject for extraction, and finally extract the required data and remove the redundant data that has not been extracted.

[0085] In the embodiment of the present invention, the steps of establishing the error compensation model are:

[0086] B1. First, scale the camera module parameters, the chip placement module parameters, and the size and position parameters of the wafer and chip to form a three-dimensional virtual operation model;

[0087] B2. Then, the operating principle of the camera module and the suction and placement module is combined with the three-dimensional virtual operation model to realize dynamic operation, and then the real-time temperature data is introduced into the three-dimensional virtual operation model to become an error compensation model;

[0088] B3. The processed operating parameter data set and image data set are transmitted to the error compensation model for analysis.

[0089] Among them, the error compensation model is generated by a system using an adaptive error correction method, which realizes a miniaturized simulation of the actual equipment's operation and simulated operation, combines the introduction of parameters for analysis and rehearsal operations, and then generates instructions for transmission based on the adjustment results after analysis.

[0090] In the embodiment of the present invention, the steps of performing the analysis operation by the error compensation model in A2 are:

[0091] C1. Introduce the operating parameter data set and the image data set into the error compensation model for analysis;

[0092] C2, realize the return operation after the first wafer pick-up and placement, and analyze the camera captured images during the return positioning;

[0093] C3, by recording the running track of the suction and placement module, and calculating the running error generated;

[0094] C4. Compensation instructions are formed based on the numerical results of the error calculation to implement correction operations on the camera module and the pickup and placement module.

[0095] In the embodiment of the present invention, the analysis operation of the collected image of the return positioning in C2 is:

[0096] c21, graying the collected image data, and then dividing the characteristic parts of the image data according to the gray value;

[0097] c22. First extract the welding head features after the previous chip is welded and sent out, and then extract the chip features after the next chip is sucked into the welding position, compare the center coordinates of the welding head features and the chip features, and determine the total error value.

[0098] In the embodiment of the present invention, the coordinate comparison operation of the chip feature and the welding head feature in c22 is:

[0099] D1. Take the center position of the welding head feature as the coordinate center, establish the X-axis in the left and right directions of the welding head feature, and establish the Y-axis in the front and rear directions of the X-axis of the welding head feature;

[0100] D2. Draw cross lines at the four corner points of the chip feature, and the intersection of the lines is the center point of the chip feature. The center point is located in the XY coordinate system in D1 and is marked as Q. The coordinate is (x 1 ,y 1 ), and the center point Q is located in the third quadrant of the XY coordinate system;

[0101] D3, and the deviation distance between the center point Q and the characteristic center of the welding head is |x 1 |, the front and rear deviation distance of the center point Q from the welding head feature center is |y 1 |, while the left and right deviation distances and the front and rear deviation distances include the deviation of the welding head feature center position affected by external factors and the inertia deviation during operation.

[0102] The collected image data is gray-scaled and then the characteristic parts in the image data are divided according to the gray-scale value. The welding head features after the previous chip is welded and sent out are extracted, and then the chip features after the next chip is absorbed to the welding position are extracted. The center coordinates of the welding head features and the chip features are compared to determine the total error value. In this way, the error of the welding head features caused by various influences can be determined, thereby realizing the differentiation and correction of the deviation of the center position of the welding head features affected by external factors and the inertial deviation in the running operation, realizing adjustment operations for different distances or situations, and ensuring the accuracy of the adjusted data.

[0103] Figure 4 In the figure, the dotted line represents the theoretical position of the welding head, and its axis is aligned with the center point of the chip. However, due to the deformation error of the downward-looking camera on the welding head, the position of the welding head will deviate from the center point of the chip when it recognizes the chip, which corresponds to the welding head of the solid line in the figure. When the welding head features are extracted first, the chip mounted under the welding head has been transported away, and there is no obstruction below. The upward-looking camera can directly capture the welding head, but when the chip is sucked to this position, the chip obstructs the welding head, so only the chip features can be extracted, and its center point represents the coordinates where the welding head should be. By comparing it with the center coordinates of the welding head feature recognition extracted last time, the error of a stroke can be determined.

[0104] In the embodiment of the present invention, the calculation operation of the running error in C3 is:

[0105] c31, determine the running track of the suction and placement module, that is, the left and right side movement operation and the front and back side movement operation of the suction and placement module when performing the return operation;

[0106] c32. Calculate the inertia deviation in the left and right directions and the inertia deviation in the front and back directions through the operating parameters of the suction and placement module.

[0107] In the embodiment of the present invention, the inertia deviation calculation operation in c32 is:

[0108] The calculation formula for the inertia deviation in the left and right directions is:

[0109] S 1 =(v 1 ×t 1 ) / 2;

[0110] And S 1 is the distance value of the inertial deviation in the left and right directions, v 1 is the inertia speed after braking to a specified distance, t 1 The time from the start of braking for the left and right side movement of the suction and placement module to the stop of the movement of the suction and placement module;

[0111] v 1 =P 1 / (F 牵 -F 制 );

[0112] P 1 In order to achieve the uniform movement of the suction and placement module, the driving power, F 牵 F is the traction force of the suction mount module during normal movement. 制 The braking force is used to stop the suction piece placement module, and the braking force is opposite to the traction force;

[0113] F 牵 =P 1 / v 2 ;

[0114] v 2 The speed at which the suction and placement module is pulled;

[0115] The calculation formula for the inertia deviation in the front and rear directions is:

[0116] S 2 =(v 3 ×t 3 ) / 2;

[0117] And S 2 is the distance value of the inertial deviation in the front and rear directions, v 3 is the inertia speed after braking to a specified distance, t 3 It is the time from the start of braking operation for the front and rear side movement of the suction and placement module to the stop of the movement of the suction and placement module.

[0118] In the embodiment of the present invention, the deviation of the characteristic center position of the welding head in D3 affected by external factors is:

[0119] The characteristic center position of the welding head is located at the left and right side deviation distance:

[0120] L 1 =S 1 +|x 1 |;

[0121] That is, adjust the left and right directions of the welding head feature center position, and compensate the deviation distance to the error compensation model to move L to the left. 1 distance;

[0122] The characteristic center position of the welding head is located in the front and rear direction. The deviation distance is:

[0123] L 2 =|y 1 |-S 2 ;

[0124] That is, the characteristic center position of the welding head is adjusted in the front and rear directions, and the deviation distance is compensated to move the error compensation model to the rear side by L 2 distance.

[0125] In the embodiment of the present invention, the operation of calculating the error at different temperatures in A3 is:

[0126] E1. Determine the temperature range to be measured, the interval is [T min , T max ], and realize the collection of operating parameters. In this temperature range, select multiple temperature points marked as T i (i=1, 2, ..., n), and measure the placement error A of the placement machine at each temperature point i , and each temperature point T i And the corresponding error value A i Record it and form a data set {(T 1 , A 1 ), (T 2 , A 2 ),…,(T n , A n )};

[0127] E2. Establish a linear model between temperature change and error, and obtain the fitting function by solving it using the least squares method;

[0128] The relationship between the linear model error A and temperature T is:

[0129] A=β 0 +β 1 T;

[0130] Among them, β 0 and β 1 are the model parameters to be determined;

[0131] And construct the error sum of squares P:

[0132] ;

[0133] In order to find the β that minimizes P 0 and β 1 , find β for P 0 and β 1 The partial derivative of , and set the partial derivative equal to 0:

[0134] ;

[0135] ;

[0136] Arrange the two equations into a linear system and solve them to obtain β 0 and β 1 :

[0137] ;

[0138] E3. Use the obtained β 0 and β 1 The predicted error values ​​of the model at different temperatures are substituted into the relationship formula of the linear model, and compared with the actual measured values ​​to verify the operation of the model.

[0139] Among them, using the obtained linear model A=β 0 +β 1 T, the placement error A at any temperature T can be predicted.

[0140] In the embodiment of the present invention, the operation of adjusting the suction and placement module braking in A2 is: compensating the inertia deviation to the suction and placement module of the error compensation model, so that the difference S at the left and right sides of the center point Q is 1 The braking operation is performed when the difference S is located in the front and rear directions from the center point Q. 2 Braking operation is performed immediately.

[0141] By determining the operation trajectory of the sheet suction and placement module, that is, the left and right side movement operation and the front and rear side movement operation of the sheet suction and placement module during the return operation, the inertia deviation in the left and right directions and the front and rear directions are calculated through the operation parameters of the sheet suction and placement module, so as to eliminate the inertia deviation by early braking, and then adjust the center position of the welding head of the sheet suction and placement module to avoid the influence of the inertia deviation on the adjustment accuracy of the center position of the welding head, so as to make the error adjustment more efficient and improve the adaptability and operation stability in the sheet picking process.

[0142] The difference between the second embodiment and the first embodiment is that the error correction method in the existing wafer fetching process and the adaptive error correction method of the present invention are used to implement the corresponding wafer fetching and mounting operations on the same wafer fetching device, and the mounting accuracy and the efficiency of completing the mounting are recorded. The specific results are shown in Table 1:

[0143] Table 1 Record results table

[0144]

[0145] To sum up, the adaptive error correction method of the present invention is applied to the wafer picking device, which has a higher accuracy in completing the mounting operation, and the time taken to complete the mounting and the average time taken to mount a single chip are shorter. Therefore, it can be better applied to actual operations and is suitable for more wafer picking devices to operate.

[0146] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0147] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0148] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An adaptive error correction method in a wafer fetching process, characterized in that: The specific steps include: A1. Use the camera module to collect parameter data during wafer pickup, and use the wafer suction and placement module to pick up the wafer and transmit the operating parameters, while the camera module and the placement module keep moving synchronously. A2. Receive the collected data and perform preliminary processing, establish an error compensation model through equipment parameters, and then introduce the processed data into the error compensation model for analysis and operation. Combined with the operation error of the placement module and the positioning error of the camera module affected by temperature, the adjustment result data of the placement module braking and the camera welding head center reference are obtained; A3. Generate correction instructions based on the result data and transmit them to the camera module and the suction and placement module to implement the adjustment operation. By adjusting the working temperature, the error at different temperatures is repeatedly calculated to achieve predictive adjustment during application, and the data generated by the operation is displayed on the display panel; The steps of analyzing the error compensation model in A2 are: C1. Introduce the operating parameter data set and the image data set into the error compensation model for analysis; C2, realize the return operation after the first wafer pick-up and placement, and analyze the camera captured images during the return positioning; C3, by recording the running track of the suction and placement module, and calculating the running error generated; C4, based on the numerical results of the error calculation, a compensation instruction is formed to implement the correction operation of the camera module and the pickup and placement module; The analysis operation of the collected image for return positioning in C2 is as follows: c21, graying the collected image data, and then dividing the characteristic parts of the image data according to the gray value; c22. First extract the welding head features after the previous chip is welded and sent out, then extract the chip features after the next chip is sucked to the position to be welded, compare the center coordinates of the welding head features and the chip features, and determine the total error value; The coordinate comparison operation of the chip feature and the welding head feature in c22 is: D1. Take the center position of the welding head feature as the coordinate center, establish the X-axis in the left and right directions of the welding head feature, and establish the Y-axis in the front and rear directions of the X-axis of the welding head feature; D2. Draw cross lines at the four corner points of the chip feature, and the intersection of the lines is the center point of the chip feature. The center point is located in the XY coordinate system in the D1 operation and is marked as Q. The coordinates are (x1, y1), and the center point Q is located in the third quadrant of the XY coordinate system. D3, and the left and right deviation distance of the center point Q from the welding head feature center is |x1|, and the front and rear deviation distance of the center point Q from the welding head feature center is |y1|, and the left and right deviation distances and the front and rear deviation distances include the deviation of the welding head feature center position affected by external factors and the inertia deviation during operation.

2. The method for adaptive error correction in a wafer fetching process according to claim 1, characterized in that: The operations of receiving the collected data and performing preliminary processing in A2 are: a21. After receiving the collected data, classify the data according to subsequent needs and perform preliminary processing of the data; a22. The data required subsequently are the operation parameter data of the suction and placement module and the image data collected by the camera module, and the redundant data is removed; a23. Summarize the data to form operating parameter data sets and image data sets.

3. The method for adaptive error correction in a wafer fetching process according to claim 1, characterized in that: The steps for establishing the error compensation model are: B1. First, scale the camera module parameters, the chip placement module parameters, and the size and position parameters of the wafer and chip to form a three-dimensional virtual operation model; B2. Then, the operating principle of the camera module and the suction and placement module is combined with the three-dimensional virtual operation model to realize dynamic operation, and then the real-time temperature data is introduced into the three-dimensional virtual operation model to become an error compensation model; B3. The processed operating parameter data set and image data set are transmitted to the error compensation model for analysis.

4. The method for adaptive error correction in a wafer fetching process according to claim 1, characterized in that: The calculation operation of the running error in C3 is: c31, determine the running track of the suction and placement module, that is, the left and right side movement operation and the front and back side movement operation of the suction and placement module when performing the return operation; c32. Calculate the inertia deviation in the left and right directions and the inertia deviation in the front and back directions through the operating parameters of the suction and placement module.

5. The method for adaptive error correction in a wafer fetching process according to claim 4, characterized in that: The inertia deviation calculation operation in c32 is: The calculation formula for the inertia deviation in the left and right directions is: S1=(v1×t1) / 2; S1 is the distance value of the inertial deviation in the left and right directions, v1 is the inertial speed after braking at the specified distance, and t1 is the time from the start of braking for the left and right side movement operation of the suction and placement module to the stop of the suction and placement module; v1=P1 / (F 牵 -F 制 ); P1 is the driving power for the suction and placement module to move at a constant speed, F 牵 F is the traction force of the suction mount module during normal movement. 制 The braking force is used to stop the suction piece placement module, and the braking force is opposite to the traction force; FAVORITE 牵 =P1 / v2; v2 is the speed at which the suction and placement module is pulled; The calculation formula for the inertia deviation in the front and rear directions is: S2 = (v3 × t3) / 2; S2 is the distance value of the inertial deviation in the front and rear directions, v3 is the inertial speed after braking at the specified distance, and t3 is the time from the start of braking for the front and rear directions of the suction and placement module to the stop of the suction and placement module.

6. The method for adaptive error correction in a wafer fetching process according to claim 5, characterized in that: The deviation of the welding head feature center position in D3 affected by external factors is: The characteristic center position of the welding head is located at the left and right side deviation distance: L1=S1+|x1|; That is, the characteristic center position of the welding head is adjusted in the left and right directions, and the deviation distance is compensated to the error compensation model moving L1 distance to the left; The characteristic center position of the welding head is located in the front and rear direction. The deviation distance is: L2=|y1|-S2; That is, the characteristic center position of the welding head is adjusted in the front and rear directions, and the deviation distance is compensated to the error compensation model moving L2 distance to the rear side.

7. The method for adaptive error correction in a wafer fetching process according to claim 5, characterized in that: The operation of calculating the error at different temperatures in A3 is: E1. Determine the temperature range to be measured, the interval is [T min , T max ], and realize the collection of operating parameters. In this temperature range, select multiple temperature points marked as T i (i=1, 2, ..., n), and measure the placement error A of the placement machine at each temperature point i , and each temperature point T i And the corresponding error value A i Record them and form a data set {(T1, A1), (T2, A2), …, (T n , A n )}; E2. Establish a linear model between temperature change and error, and obtain the fitting function by solving it using the least squares method; The relationship between the linear model error A and temperature T is: A = β0 + β1T; Among them, β0 and β1 are the model parameters to be determined; And construct the error sum of squares P: ； In order to find β0 and β1 that minimize P, we calculate the partial derivatives of β0 and β1 with respect to P and set the partial derivatives equal to 0: ； ； Arrange the two equations into a linear system and solve them to obtain β0 and β1: ； E3. Use the obtained β0 and β1 to substitute into the relationship formula of the linear model, calculate the prediction error value of the model at different temperatures, and compare it with the actual measured value to verify the operation of the model.

Images