Online monitoring system for CSP product CP test loading fixture based on internet of things

Abstract

The application relates to the technical field of wafer testing, and particularly discloses an online monitoring system of a CSP product CP test loading jig based on the Internet of Things, which is provided with an adsorption area pre-adjustment module, a CSP product coordinate determination module, a reference probe calibration module and a loading jig prediction adjustment module. The adsorption area pre-adjustment module delimits an adsorption area covered by CSP products, judges and uploads an adsorption state, pre-adjusts pressure, guarantees stable pressure before adsorption, and lays a foundation for subsequent positioning. The CSP product coordinate determination module adsorbs products in a pre-adjustment area, makes a center point enter a field of view, records initial pixel coordinates, accurately acquires position information, and provides a reference for alignment. The reference probe calibration module acquires reference probe coordinates, drives a positioning assembly to move, judges an alignment state, ensures initial alignment accuracy, and provides a basis for adjustment. The loading jig prediction adjustment module adaptively adjusts the assembly according to the alignment state, avoids deviation in advance, and improves alignment stability and accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wafer testing, in particular to an online monitoring system for a CSP product CP test loading jig based on the Internet of Things. BACKGROUND

[0002] CP testing generally refers to chip probe testing, which is an important quality verification link in the landing process of CSP products. CSP packaging technology is a packaging method that directly packages chips on a substrate without the need for pins. CSP is used in CP testing to fix, position, and achieve electrical connection with the testing equipment. The traditional CP test jig of CSP usually uses mechanical clamps or vacuum suction cups to fix the wafer, thereby fixing and positioning the wafer or chip.

[0003] For example, a semiconductor CP testing device disclosed in Chinese Patent No. CN117783817B relates to the technical field of CP testing, which includes a display and a detection box arranged on a workbench, a first electric push rod fixedly connected to the detection box, an output end of the first electric push rod fixedly connected to a probe row, a turntable feeding mechanism arranged on the right side of the detection box, a plurality of suction mechanisms for suctioning wafers arranged on the turntable feeding mechanism in a detachable manner, a linear feed mechanism arranged horizontally along the left-right direction at the bottom of the detection box, and the axis of the turntable feeding mechanism perpendicular to and intersecting the axis of the linear feed mechanism.

[0004] For example, a semiconductor CP testing system disclosed in Chinese Patent No. CN116660709A relates to the technical field of CP testing, which includes an EAP system, a personnel operation end, a CPA system, a probe station, and a testing machine. The personnel operation end is in communication connection with the CPA system for operation by the worker. The EAP system is in communication connection with the CPA system for sending Map information of the semiconductor product to the CPA system. The probe station is in communication connection with the CPA system for probe walking according to the instruction of the CPA system and sending feedback information to the CPA system. The testing machine is in communication connection with the CPA system for testing according to the instruction of the CPA system and sending the test result to the CPA system. The CPA system generates a Map graph based on the Map information and controls the probe station to walk according to the Map graph. Then, after the probe station moves to the specified position, the testing machine is notified to perform testing. Finally, the test result fed back by the testing machine is saved and analyzed.

[0005] In combination with the above technical solutions, it is found that the existing CP test chip fixing method focuses on test process automation and mechanical structure optimization, relies on static CP test monitoring, and the CP test chip fixing needs real-time feedback of the physical posture of the chip, thus causing abnormal conditions to be unable to be captured in time, for example, mechanical clamps can damage the wafer surface, and the vacuum chuck cannot adjust the platform after adsorption for slight deformation, causing product testing to be poor, etc., and the existing technology only saves and analyzes test data locally, and the static monitoring and local data processing mode cannot meet the real-time requirement, thus abnormal processing relies on post-response, and it is difficult to realize online monitoring of the adsorption state in the CP test process. SUMMARY

[0006] In view of the deficiencies of the prior art, the present application provides an online monitoring system of a CSP product CP test loading fixture based on the Internet of Things, which can effectively solve the problems involved in the background art.

[0007] To achieve the above object, the present application is implemented by the following technical solutions: an online monitoring system of a CSP product CP test loading fixture based on the Internet of Things, comprising an adsorption area pre-adjustment module, used for recording the area combined by a plurality of positioning component adsorption heads under the CSP product as a positioning component adsorption area, determining the adsorption state of the positioning component adsorption area, uploading the adsorption state to a monitoring scheduling port based on the Internet of Things, and pre-adjusting the adsorption pressure of the positioning component adsorption area; a CSP product coordinate determination module, used for performing CSP product adsorption based on the pre-adjusted positioning component adsorption area, controlling the center point of the CSP product to enter the field of view of a vision device by the positioning component, and recording the initial pixel coordinates of the center point of the CSP product by the vision device; a reference probe calibration module, used for obtaining the pixel coordinates of the reference probe, moving the positioning component to the pixel coordinates of the reference probe by a motion component, and determining the alignment state of the reference probe and the center point of the CSP product; and a loading fixture prediction adjustment module, used for adaptively predicting and adjusting the components belonging to the loading fixture according to the alignment state.

[0008] Compared with the prior art, the embodiments of the present application have at least the following advantages or beneficial effects:

[0009] (1) The present application provides an online monitoring system for CSP product CP test loading fixture based on Internet of Things, the adsorption area pre-adjustment module delimits the adsorption area covered by the CSP product, judges and uploads the adsorption state, pre-adjusts the pressure, ensures the stable pressure before adsorption, and lays the foundation for subsequent positioning; the CSP product coordinate determination module adsorbs the product in the pre-adjustment area, makes the center point enter the field of view and records the initial pixel coordinates, accurately obtains the position information, and provides the alignment reference; the reference probe calibration module obtains the reference probe coordinates, drives the positioning component to move and judges the alignment state, ensures the initial alignment accuracy, and provides the basis for adjustment; the loading fixture prediction adjustment module adjusts the component adaptively according to the alignment state, avoids deviation in advance, improves the alignment stability and accuracy. Through step-by-step optimization of the alignment effect, the subsequent operation is ensured to be smooth.

[0010] (2) The present application evaluates the operation effective parameters of the abnormal adsorption head, which can comprehensively capture the key characteristics such as pressure deviation, response time, power consumption and temperature rise of the adsorption head, accurately locate the abnormal states such as overpressure, jam and aging, etc. This not only can avoid the CSP product micro-deformation and positioning deviation caused by adsorption imbalance, or equipment failure caused by continuous abnormality, but also can provide data support for adsorption pressure correction and component maintenance, ensure that the adsorption area pressure is always in a stable and uniform state, ensure the adsorption positioning accuracy of CSP products from the source, and reduce the difficulty of subsequent alignment adjustment.

[0011] (3) The present application can not only avoid the waste of resources and time loss caused by repeated data collection, reduce the system running load, but also can build a complete adjustment link through the logical association between parameters (such as converting the deviation parameter into a compensation amount, using the adsorption pressure data to guide the pressure correction), so that the decision of each module is more coherent and scientific, and the accuracy and response speed of the adjustment strategy are greatly improved.

[0012] (4) Compared with the prior art, the process has the following advantages: the prior art relies on a single parameter or lag adjustment, the alignment accuracy is easily affected by the environment and equipment wear, and lacks full-process cooperation; the process lays a stable foundation through the pre-adjustment module, builds a closed-loop control combined with visual positioning and real-time monitoring, resolves the deviation in advance through multi-module cooperation and prediction adjustment, integrates the advantages of Internet of Things, visual identification and intelligent algorithm, realizes the full-process optimization from adsorption to alignment, effectively overcomes the problems of low accuracy, slow response and poor stability in traditional technology, and significantly improves the efficiency and reliability of CSP product testing. BRIEF DESCRIPTION OF DRAWINGS

[0013] The application will be further described with the accompanying drawings, but the embodiments in the drawings do not constitute any limitation to the application, and other embodiments can be obtained by those skilled in the art without creative labor on the basis of the following drawings.

[0014] Figure 1 It is a schematic diagram of system module connection of the application.

[0015] Figure 2 It is a brief schematic diagram of coordinate projection.

[0016] Figure 3 It is a schematic diagram of overall architecture of the loading fixture.

[0017] Figure 4 It is a detail display diagram of the angle adjustment assembly.

[0018] Figure 5 It is a detail display diagram of the positioning assembly.

[0019] Reference signs: origin a of reference probe pixel coordinate system, termination pixel coordinate b of CSP product center point, horizontal straight line d, motion assembly 1, rotation assembly 2, angle adjustment assembly 3, positioning assembly 4, loose clamping mechanism 5, mounting plate 3-01, adjustment motor 3-02, precision lead screw 3-03, adjustment motor 3-04, precision lead screw 3-05, bottom plate 4-01, support device 4-06, support device 4-07, pressing block 4-08, torsion spring 4-09, base 4-10, suction head 4-11, mounting plate 5-01, voice coil motor 5-02, and auxiliary guide rod mechanism 5-03. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the application will be clearly and completely described in combination with the accompanying drawings in the embodiments of the application.

[0021] In the embodiments of the application, the words such as "example", "for example" and the like are used to represent an example, illustration or description. Any embodiment or design scheme described as "example" in the application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. In fact, the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the application, the meaning expressed by "and / or" can be both, or can be one of the two optionally.

[0022] In the embodiments of the present application, "image" and "picture" can be used interchangeably, and it should be pointed out that the meanings expressed are consistent when the distinction is not emphasized.

[0023] In the embodiments of the present application, sometimes the subscript such as W1 can be written in the form of non-subscript such as W1, and the meanings expressed are consistent when the distinction is not emphasized.

[0024] To make the technical problems, technical solutions and advantages to be solved by the present application clearer, specific embodiments will be described in detail below with reference to the drawings.

[0025] Referring to Figure 1 The embodiments of the present application provide a technical solution: an online monitoring system of a CSP product CP test loading fixture based on the Internet of Things, comprising an adsorption area pre-adjustment module, a CSP product coordinate determination module, a reference probe calibration module, a loading fixture prediction adjustment module and a monitoring scheduling information library. The monitoring scheduling information library is used to store the pressure response average definition duration, the electromagnetic valve average definition power consumption, the surface definition temperature rise and the preset values of various parameters.

[0026] The adsorption area pre-adjustment module is connected with the CSP product coordinate determination module, the CSP product coordinate determination module is connected with the reference probe calibration module, the reference probe calibration module is connected with the loading fixture prediction adjustment module, and the adsorption area pre-adjustment module, the CSP product coordinate determination module, the reference probe calibration module and the loading fixture prediction adjustment module are all connected with the monitoring scheduling information library.

[0027] The adsorption area pre-adjustment module is used to record the area combined by the adsorption heads of a plurality of positioning components under the CSP product as a positioning component adsorption area, to determine the adsorption state of the positioning component adsorption area, to upload the adsorption state to the monitoring scheduling port based on the Internet of Things, and to pre-adjust the adsorption pressure of the positioning component adsorption area.

[0028] The CSP product involved in the embodiments of the present application specifically refers to a chip using chip scale package (Chip Scale Package) technology.

[0029] The positioning assembly adsorption area, through the synergistic effect of multiple adsorption heads, stably grabs the CSP chip in the adsorption area, ensures that it maintains an accurate spatial position in the detection link, and avoids packaging defects caused by position deviation. The area formed by the combination of multiple adsorption heads can disperse the adsorption force on the CSP product, prevent the chip from being deformed or damaged due to excessive force on a single adsorption head, and is especially suitable for thin and fragile CSP products, improving the stability of the operation process.

[0030] It should be explained that the above Internet of Things is used as the core of connecting physical devices and monitoring systems in the embodiments of the present application. By deploying sensors on the adsorption heads of the positioning assembly, the working state of each adsorption head is collected in real time, and the data is uploaded to the monitoring and scheduling port through the Internet of Things communication protocol, realizing dynamic monitoring of the adsorption process. The monitoring and scheduling port can centrally display the real-time state of all positioning assembly adsorption areas, trigger alarms through the Internet of Things, and record historical adsorption state data through the Internet of Things subsystem. Specifically, through data connection and intelligent scheduling, the embodiments of the present application can monitor, trace and optimize, ultimately improving the stability and efficiency of CSP production.

[0031] The positioning assembly is specifically as shown in Figure 5 , Figure 5 is a detail display diagram of the positioning assembly. The positioning assembly 4 is installed on the angle adjusting assembly 3 through the bottom plate 4-01. The base 4-10 is mounted on the bottom plate 4-01, and the base 4-10 is made of metal material; two supporting devices 4-06 and supports 4-07 are mounted on the base 4-10; a pressing block 4-08 is mounted on the base 4-10, and the pressing block 4-08 is driven to press the workpiece through a torsion spring 4-09. A loose clamp mechanism 5 is mounted on the base 4-10. The loose clamp mechanism 5 is installed on the bottom plate 4-01 through the mounting plate 5-01; the loose clamp mechanism 5 is powered to expand and contract through the voice coil motor 5-02, and the loose clamp mechanism 5 is provided with an auxiliary guide rod mechanism 5-03 to ensure the rigidity of expansion and contraction. The head of the loose clamp mechanism 5 is provided with a pressing block 5-03, and the pressing block 5-03 is made of non-metallic material.

[0032] The base 4-10 is internally provided with an adsorption head 4-11 made of elastic material, which realizes stable grabbing, positioning and releasing of the chip by generating negative pressure (vacuum adsorption) or positive pressure (pneumatic clamping). The core functions include: tightly adhering the workpiece to the fixture surface through uniform adsorption force to prevent positioning deviation of the workpiece caused by vibration and displacement during handling and testing; balancing the stress on the workpiece surface through the synergistic effect of multiple adsorption heads (such as array layout) to avoid workpiece deformation (such as wafer warping and CSP package cracking) caused by local overpressure; reliably grabbing different sizes and materials of workpieces (such as thin wafers and flexible substrates) by adjusting the adsorption pressure and replacing adsorption heads of different aperture / materials (such as silica gel suction cups). A micro pressure sensor is embedded at the connection between the adsorption head and the vacuum pipeline to collect the adsorption pressure value in real time, which is analogous to the test point for monitoring voltage on the circuit board.

[0033] Specifically, the adsorption state of the adsorption area of the positioning assembly is determined, and the specific determination process is as follows:

[0034] The initial adsorption pressure of a plurality of adsorption heads of the positioning assembly covered by the adsorption area of the positioning assembly is obtained, wherein the adsorption pressure can be extracted from the running record of the vacuum generator, and is recorded as the initial adsorption pressure of each adsorption head. The initial adsorption pressure of each adsorption head is processed by variance to obtain the adsorption pressure uniformity of the adsorption area of the positioning assembly.

[0035] The adsorption pressure uniformity of the adsorption area of the positioning assembly is compared with the pre-defined adsorption pressure uniformity threshold to obtain an adsorption pressure comparison result, which includes an adsorption pressure comparison first result and an adsorption pressure comparison second result.

[0036] The adsorption pressure comparison first result indicates that the adsorption pressure uniformity of the adsorption area of the positioning assembly is less than or equal to the adsorption pressure uniformity threshold, and the adsorption pressure comparison second result indicates that the adsorption pressure uniformity of the adsorption area of the positioning assembly is greater than the adsorption pressure uniformity threshold.

[0037] If the adsorption pressure comparison result shows the adsorption pressure comparison first result, it is determined that the adsorption state of the adsorption area of the positioning assembly is uniform adsorption state, and the adsorption pressure of the adsorption area of the positioning assembly does not need to be pre-adjusted. The adsorption area of the positioning assembly performs CSP product adsorption.

[0038] If the adsorption pressure comparison result shows the adsorption pressure comparison second result, it is determined that the adsorption state of the adsorption area of the positioning assembly is non-uniform adsorption state, and the adsorption pressure of the adsorption area of the positioning assembly needs to be corrected.

[0039] After the adsorption pressure correction is completed, the adsorption state of the adsorption area of the positioning assembly is re-evaluated to determine whether iterative correction is needed.

[0040] Further, the adsorption pressure correction is performed on the adsorption area of the positioning assembly, and the specific correction process is as follows:

[0041] The average adsorption pressure of the adsorption area of the positioning assembly is extracted, and the absolute difference value processing is performed on the average adsorption pressure of the adsorption area of the positioning assembly and the initial adsorption pressure of each adsorption head, respectively, to obtain the initial adsorption pressure deviation of each adsorption head, and the initial adsorption pressure deviation of each adsorption head is arranged in descending order to obtain the initial adsorption pressure deviation sequence of each adsorption head, and the adsorption head corresponding to the first arrangement order is recorded as an abnormal adsorption head.

[0042] The above iteration correction is specifically that if the re-evaluation of the adsorption state of the adsorption area of the positioning assembly is still the uneven adsorption state, the adsorption head corresponding to the second initial initial adsorption pressure deviation is recorded as an abnormal adsorption head, and the adsorption pressure correction is performed.

[0043] The pre-adjustment running state information of the abnormal adsorption head is obtained, the running effective parameter of the abnormal adsorption head is evaluated, the running effective parameter of the abnormal adsorption head is compared with the pre-defined running effective parameter threshold, if the running effective parameter of the abnormal adsorption head is greater than or equal to the running effective parameter threshold, the monitoring scheduling port sends the adsorption pressure adjustment instruction to the positioning assembly to compensate for the initial adsorption pressure deviation of the abnormal adsorption head. It needs to be explained that the initial adsorption pressure deviation in the embodiment represents the adsorption pressure deviation when the absolute value of the first initial adsorption pressure deviation is positive.

[0044] If the running effective parameter of the abnormal adsorption head is less than the running effective parameter threshold, the abnormal adsorption head is isolated, and a compensation adsorption group is constructed to compensate for the missing adsorption pressure of the abnormal adsorption head.

[0045] Specifically, the compensation adsorption group is constructed to compensate for the missing adsorption pressure of the abnormal adsorption head, and the specific construction process is as follows:

[0046] The straight line distance from the abnormal adsorption head to the center point of the adsorption area of the positioning assembly is obtained, wherein the straight line distance can be extracted in the pixel coordinate system of the positioning assembly, and is recorded as the center distribution distance of the abnormal adsorption head. The center distribution distance of the abnormal adsorption head is matched with the pre-defined center distribution distance interval, if the center distribution distance of the abnormal adsorption head belongs to the center distribution distance interval, it is determined that the abnormal adsorption head belongs to the center area, otherwise it is determined that the abnormal adsorption head belongs to the edge area.

[0047] If the abnormal adsorption head belongs to the center area, the center area is obtained. A plurality of normal adsorption heads in the center area are obtained, the average adsorption pressure of the adsorption area of the positioning assembly is evenly distributed to the plurality of normal adsorption heads in the center area, and a compensation adsorption group is formed to compensate for the missing adsorption pressure of the abnormal adsorption head.

[0048] The above average adsorption pressure of the adsorption area of the positioning assembly includes the missing adsorption pressure of the abnormal adsorption head.

[0049] If the abnormal adsorption head belongs to the edge area, the center distribution distance of the abnormal adsorption head is subtracted from the maximum value of the center distribution distance interval to obtain the center deviation distance of the abnormal adsorption head, and the reference number of the adsorption head of the compensation adsorption group is mapped. According to the adsorption pressure uniformity of the positioning component adsorption area, the adsorption head number correction amount is mapped to form a coupling relationship with the reference number of the adsorption head of the compensation adsorption group. Specifically, the adsorption head number correction amount is multiplied by the reference number of the adsorption head of the compensation adsorption group to obtain the adaptive number of the adsorption head of the compensation adsorption group.

[0050] The above mapping obtains the reference number of the adsorption head of the compensation adsorption group. Specifically, the mapping set between the center deviation distance and the reference number of the adsorption head is obtained from the monitoring scheduling information library. The real-time center deviation distance of the abnormal adsorption head is brought into the mapping set to obtain the reference number of the adsorption head of the compensation adsorption group.

[0051] The greater the center deviation distance, the wider the influence range of the abnormal adsorption head on the overall uniformity of the adsorption area, and the more the reference number of the adsorption head is needed to offset the local abnormality through more peripheral normal adsorption heads. Conversely, the smaller the center deviation distance, the more limited the influence range, and the fewer the reference number. Through this mapping relationship, the number of compensation adsorption heads can be dynamically matched according to the actual range of abnormal influence, avoiding insufficient compensation caused by insufficient reference number (such as local adsorption imbalance), and preventing resource waste or excessive compensation caused by excessive reference number (such as causing new adsorption imbalance), thereby accurately ensuring the pressure uniformity of the adsorption area, stabilizing the wafer attitude, and improving the reliability and efficiency of CP testing.

[0052] The above mapping obtains the adsorption head number correction amount. Specifically, the mapping set between the adsorption pressure uniformity and the adsorption head number correction amount is obtained from the monitoring scheduling information library. The real-time adsorption pressure uniformity of the positioning component adsorption area is brought into the mapping set to obtain the adsorption head number correction amount.

[0053] The lower the adsorption pressure uniformity (the greater the difference between the adsorption heads), the greater the adsorption head number correction amount required, that is, more adsorption heads are needed to participate in pressure regulation to balance the local deviation. The higher the adsorption pressure uniformity (the more consistent the pressure distribution), the smaller the correction amount, and only a small number of adsorption heads are needed to fine-tune to maintain stability. Through this mapping, the number of adsorption heads participating in regulation can be dynamically optimized according to the actual pressure distribution state, avoiding the lack of adsorption heads leading to insufficient uniformity improvement, and preventing the intervention of too many adsorption heads causing regulation redundancy and increased energy consumption. Thus, the pressure balancing capability of the adsorption system is accurately improved, the wafer is stably fixed, the test deviation caused by uneven stress is reduced, and the accuracy and efficiency of CP testing are ultimately improved.

[0054] The compensation adsorption group compensates for the missing adsorption pressure of the abnormal adsorption head.

[0055] Further, the operation effective parameter of the abnormal adsorption head is evaluated, and the specific evaluation process is as follows:

[0056] The pre-adjustment operation state information of the abnormal adsorption head includes the average pressure response time length of the abnormal adsorption head, the average electromagnetic valve power consumption of the abnormal adsorption head, and the surface temperature rise of the abnormal adsorption head in the pre-adjustment period, wherein the operation state information can be extracted from the operation record of the monitoring point device to which the abnormal adsorption head belongs.

[0057] It should be explained that the surface temperature rise mentioned above refers to the change rate of the surface temperature of the adsorption head with the running time.

[0058] The average pressure response time length, the average electromagnetic valve power consumption, and the surface temperature rise are obtained from the monitoring scheduling information library.

[0059] The statistical deviation processing result includes the deviation processing result between the average pressure response time length of the abnormal adsorption head and the average pressure response time length, the deviation processing result between the average electromagnetic valve power consumption of the abnormal adsorption head and the average electromagnetic valve power consumption, and the deviation processing result between the surface temperature rise of the abnormal adsorption head and the surface temperature rise. The influence factor and the deviation processing result are sequentially weighted and aggregated to obtain the operation effective parameter of the abnormal adsorption head, and the specific analysis process is as follows:

[0060]

[0061] In the formula, WE is the operation effective parameter of the abnormal adsorption head, MT is the average pressure response time length of the abnormal adsorption head, MT' is the average pressure response time length, PC is the average electromagnetic valve power consumption of the abnormal adsorption head, PC' is the average electromagnetic valve power consumption, TR is the surface temperature rise of the abnormal adsorption head, TR' is the surface temperature rise, d1 is the influence factor corresponding to the average pressure response time length in the monitoring scheduling information library, d2 is the influence factor corresponding to the average electromagnetic valve power consumption in the monitoring scheduling information library, and d3 is the influence factor corresponding to the surface temperature rise in the monitoring scheduling information library.

[0062] The influence factor corresponding to the average duration of pressure response, the influence factor corresponding to the average power consumption of the electromagnetic valve, and the influence factor corresponding to the surface temperature rise are all obtained from the monitoring scheduling information library. For example, the average duration of pressure response, the average power consumption of the electromagnetic valve, and the surface temperature rise form a mapping set with the influence factor corresponding to the average duration of pressure response, the influence factor corresponding to the average power consumption of the electromagnetic valve, and the influence factor corresponding to the surface temperature rise preset in the monitoring scheduling information library, respectively. The real-time average duration of pressure response, the average power consumption of the electromagnetic valve, and the surface temperature rise are brought into the mapping set to obtain the influence factor corresponding to the average duration of pressure response, the influence factor corresponding to the average power consumption of the electromagnetic valve, and the influence factor corresponding to the surface temperature rise.

[0063] In this embodiment, there is a significant chain effect among the average duration of pressure response of the abnormal adsorption head, the average power consumption of the electromagnetic valve, and the surface temperature rise: an increase in the average power consumption of the electromagnetic valve (such as an increase in current due to valve core sticking or coil aging) will directly exacerbate heating, causing the surface temperature to rise; a high temperature environment will further increase the coil resistance, causing the power consumption to further climb, and the valve core expansion caused by heating may exacerbate the sticking, prolonging the average duration of pressure response, all of which will greatly reduce the operational effectiveness of the abnormal adsorption head; conversely, a too long pressure response duration means that the electromagnetic valve needs to work frequently or continuously to adjust the pressure, which will accumulate higher power consumption and push up the surface temperature, collectively reflecting potential failures such as mechanical sticking and electrical performance degradation of the adsorption head.

[0064] The CSP product coordinate determination module is configured to perform CSP product adsorption based on the pre-adjusted positioning assembly adsorption area, the positioning assembly controls the center point of the CSP product to enter the field of view of the vision device, and the vision device records the initial pixel coordinates of the center point of the CSP product.

[0065] The reference probe calibration module is configured to obtain the pixel coordinates of the reference probe, and drive the positioning assembly to move towards the pixel coordinates of the reference probe through the motion assembly, and determine the alignment state of the reference probe and the center point of the CSP product.

[0066] Specifically, the motion assembly drives the positioning assembly to move towards the pixel coordinates of the reference probe, and the specific movement process is as follows:

[0067] According to the pixel coordinates of the reference probe, a reference probe pixel coordinate system is constructed, and according to the initial pixel coordinates of the center point of the CSP product, a CSP product pixel coordinate system is constructed.

[0068] The above-mentioned construction of the reference probe pixel coordinate system is specifically as follows: the reference probe tip is located by a vision device, its pixel coordinates in the camera's field of view are recorded, the X / Y axis of the motion component is moved so that the reference probe tip is aligned with the center of the camera's field of view, the mechanical coordinates of the motion component at this time are recorded, and this is used as the origin (or reference point) of the reference probe pixel coordinate system. The X / Y axis direction of the motion component is used as the direction of the reference probe pixel coordinate system, and the reference probe pixel coordinate system is constructed in this way.

[0069] The aforementioned construction of the CSP product pixel coordinate system specifically involves the positioning component bringing the center point of the CSP product into the field of view of the vision device. The center point of the CSP product refers to the central crosshair target. The vision device captures an image of the CSP product, extracts the central crosshair target, and records it in the pixel image to obtain the pixel coordinates of the CSP product's center point. A preset marker point is selected on the edge of the CSP product. This preset marker point is defined as a point aligned with the horizontal direction of the reference probe pixel coordinate system and lying on the same straight line as the center point. It is used to determine the X / Y axis direction of the CSP product's pixel coordinates, with the straight line from the center point to the marker point serving as the X-axis direction.

[0070] The angle between the X-axis of the CSP product pixel coordinate system and the X-axis of the reference probe pixel coordinate system is denoted as the rotation angle. The offset of the origin of the CSP product pixel coordinate system relative to the origin of the reference probe pixel coordinate system in the X-axis direction is denoted as the X translation amount. The offset of the origin of the CSP product pixel coordinate system relative to the origin of the reference probe pixel coordinate system in the Y-axis direction is denoted as the Y translation amount.

[0071] The rotation angle is specifically the angle by which the CSP product's pixel coordinate system rotates along the X-axis towards the X-axis of the reference probe's pixel coordinate system. It should be noted that the rotation angle in this embodiment is based on angles. accomplish.

[0072] A mapping matrix is ​​formed based on the rotation angle, X translation amount, and Y translation amount. Based on the mapping matrix, the motion component drives the positioning component to move towards the pixel coordinates of the reference probe.

[0073] The above-mentioned motion components are specifically as follows: Figure 3 As shown, Figure 3 This is a schematic diagram of the overall architecture of the loading fixture, including an XY motion component 1, a rotation component 2, an angle adjustment component 3, and a positioning component 4. The XY motion component 1 consists of two orthogonal high-precision linear motors, which drive the motion along the x and y directions. The motion component is the core execution unit for achieving precise displacement and positioning in automated processes such as CSP product testing. It is mainly responsible for driving components such as the positioning component, the suction head, or the loading fixture to complete linear, rotary, or multi-axis composite movements to meet the position adjustment requirements of the workpiece during suction, alignment, and transfer.

[0074] The above mapping matrix is essentially to establish a conversion relationship between the physical coordinates of the positioning assembly and the pixel coordinates of the reference probe by the mapping matrix, to convert the pixel position of the reference probe in the image into physical displacement parameters (such as millimeter-level movement amount of X and Y axes) executable by the motion assembly, and then to drive the positioning assembly to move according to these parameters, so that the adsorption area or test station of the positioning assembly and the pixel coordinates of the reference probe are accurately overlapped in the physical space. This process eliminates the scale difference, coordinate system offset and other errors between the image pixels and the physical space through the mapping matrix, realizes the direct conversion of visual positioning information to mechanical motion instructions, ensures the high-precision alignment of the positioning assembly and the reference probe, and provides a motion control basis for the accurate contact between the probe and the chip pad in the subsequent CP test.

[0075] Further, the alignment state of the reference probe and the center point of the CSP product is determined, and the specific analysis process is as follows:

[0076] The motion assembly drives the positioning assembly to move towards the pixel coordinates of the reference probe, obtains the terminal pixel coordinates of the center point of the CSP product, and calibrates the pixel coordinates of the reference probe. The specific process of calibration is as follows:

[0077] It should be explained that in the process of calibrating the terminal pixel coordinates of the center point of the CSP product with the pixel coordinates of the reference probe, if the positioning assembly where the CSP product is located is inclined at an angle, the horizontal adjustment is performed based on the adjustment motor and the precision lead screw of the angle adjustment assembly. Specifically, the horizontal adjustment is realized based on the angle and the angle The alignment of the reference probe and the center point of the CSP product is performed on the horizontal plane.

[0078] The terminal pixel coordinates of the center point of the CSP product and the pixel coordinates of the reference probe are subjected to vector processing to obtain the alignment deviation of the probe-CSP product.

[0079] The above vector processing is to calculate the straight-line distance between the terminal pixel coordinates and the pixel coordinates by the Euclidean distance.

[0080] According to the adsorption pressure uniformity of the adsorption area of the positioning assembly, the alignment deviation adjustment factor is obtained by mapping, coupled with the preset alignment deviation to obtain the alignment defined deviation.

[0081] The above mapping of the alignment deviation adjustment factor is specifically to obtain the mapping set between the adsorption pressure uniformity and the alignment deviation adjustment factor from the monitoring and scheduling information library, to obtain the alignment deviation adjustment factor by bringing the adsorption pressure uniformity of the adsorption area of the positioning assembly into the mapping set.

[0082] The lower the adsorption pressure uniformity (the more uneven the pressure distribution), the greater the alignment deviation adjustment factor, and the more significant mechanical adjustment (such as screw displacement, rotation angle compensation) is needed to offset the micro-deformation deviation of the wafer caused by uneven force; the higher the uniformity (the more balanced the pressure distribution), the smaller the adjustment factor, and only fine tuning is needed to maintain the alignment accuracy. Through this mapping, the abstract pressure uniformity index can be converted into specific alignment adjustment parameters, which not only avoids the accumulation of alignment deviation caused by insufficient adjustment when the uniformity is insufficient, but also prevents mechanical vibration or positioning overshoot caused by excessive adjustment, thereby dynamically ensuring the relative position accuracy of the wafer and the probe, and improving the stability and data reliability of the CP test.

[0083] The alignment deviation of the probe-CSP product is compared with the alignment defined deviation to obtain an alignment deviation comparison result, and the alignment deviation comparison result includes an alignment deviation first comparison result and an alignment deviation second comparison result.

[0084] The alignment deviation first comparison result indicates that the alignment deviation of the probe-CSP product is less than or equal to the alignment defined deviation, and the alignment deviation second comparison result indicates that the alignment deviation of the probe-CSP product is greater than the alignment defined deviation.

[0085] If the alignment deviation comparison result is the alignment deviation first comparison result, it is determined that the alignment state of the probe-CSP product is calibration matching, and mechanical adjustment is not needed to be performed, and if the alignment deviation comparison result is the alignment deviation second comparison result, it is determined that the alignment state of the probe-CSP product is calibration mismatching, and mechanical adjustment is performed.

[0086] Specifically, mechanical adjustment is performed, and the specific analysis process is as follows:

[0087] The data processing unit to which the visual equipment belongs projects the terminal pixel coordinates of the center point of the CSP product in the reference probe pixel coordinate system along the X-axis direction, wherein the included angle between the X-axis of the CSP product terminal pixel coordinate system and the X-axis of the reference probe pixel coordinate system is based on the terminal pixel coordinates of the center point of the CSP product, and the CSP product terminal pixel coordinate system is represented as a coordinate system re-established based on the terminal pixel coordinates of the center point of the CSP product, to obtain the compensation rotation angle of the probe-CSP product. The compensation rotation angle is based on the angle is realized.

[0088] The specific projection process is as shown in Figure 2 . Figure 2For the coordinate projection schematic diagram, a represents the origin of the reference probe pixel coordinate system, b is the terminal pixel coordinate of the CSP product center point, the projection is made along the X axis direction in the reference probe pixel coordinate system, the included angle between the horizontal straight line d and the X axis of the terminal pixel coordinate system is denoted as the compensation rotation angle, the horizontal straight line d is parallel to the X axis of the reference probe pixel coordinate system, and the rotation center of the compensation rotation angle is the center point of the CSP product. Because in the alignment process, the CSP product posture needs to be adjusted by rotating around the center point itself (avoiding the center point position offset during rotation), it is ensured that the relative position relationship between the CSP product and the reference probe after rotation is only corrected by an angle, rather than introducing additional translation deviation. The projection value of the X axis in the reference probe pixel coordinate system is X compensation translation, and the projection value of the Y axis in the reference probe pixel coordinate system is Y compensation translation.

[0089] The projection value of the X axis of the CSP product center point in the reference probe pixel coordinate system is denoted as the X compensation translation of the CSP product center point, and the projection value of the Y axis of the CSP product center point in the reference probe pixel coordinate system is denoted as the Y compensation translation of the CSP product center point.

[0090] According to the compensation rotation angle, the X compensation translation and the Y compensation translation, a compensation mapping matrix is formed, the compensation mapping matrix is uploaded to the motion component based on the Internet of Things, a mechanical adjustment instruction is generated by a monitoring and scheduling port, and the motion component receives the mechanical adjustment instruction to drive the positioning component to move to the pixel coordinate of the reference probe again according to the compensation mapping matrix.

[0091] The rotation angle is executed by the angle adjustment component connected by the rotation component, and the X compensation translation and the Y compensation translation are executed by the motion component.

[0092] It needs to be explained that the rotation component 2 is powered by a high-precision DD motor, so that it can rotate along the Z axis and adjust the angle θ, as shown in Figure 3 The angle adjustment on the X axis is realized according to the angle , the angle adjustment on the Y axis is realized according to the angle , and the angle adjustment on the Z axis is realized according to the angle .

[0093] The angle adjustment component specifically Figure 4 as shown, Figure 4For the angle adjustment assembly detail show map, the angle adjustment assembly 3 is installed on the rotating assembly 2 through the mounting plate 3-01. The mounting plate 3-01 is provided with an adjustment motor 3-02, and the adjustment motor 3-02 is connected with a precision lead screw 3-03. Under the driving of the adjustment motor 3-02, the angle a is adjusted through the action of the precision lead screw. Specifically, the lead screw drives a movable sleeve that is sleeved on the lead screw and meshes with the lead screw to move. The movable sleeve (not shown in the figure) drives the platform to move along the fixed guide rail (not shown in the figure) through the connecting rod assembly (not shown in the figure). The guide rail (not shown in the figure) is arranged in an arc shape, so that the angle adjustment can be realized. In addition, the platform is driven to move in a circular arc by the motor. There are many internal transmission forms, such as using a lead screw and an arc gear (not shown in the figure) to move, and the arc gear drives the platform to adjust the angle. The adjustment motor 3-04 is connected with the precision lead screw. Under the driving of the adjustment motor 3-04, the angle β is adjusted through the action of the precision lead screw 3-05. The principle of angle adjustment of the adjustment motor 3-04 is the same as that of the aforementioned adjustment motor 3-02. In the specific use process, a two-dimensionally adjustable electric angle positioner on the market can be selected as a module for assembly.

[0094] The loading jig prediction adjustment module is used for adaptive prediction adjustment of components belonging to the loading jig according to the alignment state.

[0095] Further, the components belonging to the loading jig are adaptively prediction adjusted according to the alignment state, and the specific adjustment process is as follows:

[0096] The F-norms of the mapping matrix and the compensation mapping matrix are calculated respectively to obtain the characteristic evaluation value of the mapping matrix and the characteristic evaluation value of the compensation mapping matrix respectively. The characteristic evaluation value of the compensation mapping matrix is processed by ratio with the characteristic evaluation value of the mapping matrix to obtain the proportion deviation value of the probe-CSP product.

[0097] It needs to be explained that the core beneficial effect of the above-mentioned proportion deviation value of the probe-CSP product is that the relative level of quantitative adjustment accuracy: the smaller the ratio, the lower the proportion of the characteristic evaluation value of the compensation mapping matrix (reflecting the overall size of the compensation amount) relative to the characteristic evaluation value of the mapping matrix (reflecting the overall size of the initial adjustment amount), which means that only a small compensation amount is needed to offset the deviation on the basis of the initial adjustment, indicating that the initial movement control accuracy of the system is high, the compensation mechanism is accurate and efficient, and the adjustment accuracy is higher. On the contrary, the larger the ratio, the more significant the initial adjustment deviation, and a larger compensation amount is needed to correct, and the adjustment accuracy is relatively low. Through this quantitative index, the overall accuracy performance of the system from the initial adjustment to the compensation correction can be directly judged, providing clear numerical basis for optimizing control parameters and improving alignment stability, avoiding fuzzy judgment of the adjustment effect, and making the accuracy evaluation more objective and operable.

[0098] The probe-CSP product proportion deviation value is compared with predefined proportion deviation intervals, including a first proportion deviation interval, a second proportion deviation interval, and a third proportion deviation interval.

[0099] It should be explained that the first proportion deviation interval, the second proportion deviation interval, and the third proportion deviation interval are set, the maximum threshold of the first proportion deviation interval is T1, the maximum threshold of the second proportion deviation interval is T2, and the maximum threshold of the third proportion deviation interval is T3, and 0 < T1 < T2 < T3.

[0100] If the probe-CSP product proportion deviation value belongs to the first proportion deviation interval, the probe-CSP product proportion deviation value is written to the monitoring scheduling port as a compensation amount for the next round of alignment operation.

[0101] In this embodiment, the deviation data is used as the core compensation parameter for the next round of alignment operation, and the monitoring system automatically calculates and generates targeted adjustment instructions (such as probe position fine adjustment amount and CSP product carrier displacement compensation value) according to this value, to ensure that the probe can accurately correct the previous deviation in the next alignment operation, and realize high-precision fitting with the CSP product test points. This process forms a closed-loop control of detection-feedback-compensation, which continuously reduces the alignment error by real-time transmission of deviation information and conversion into compensation actions, significantly improves the consistency and yield of CSP product testing, reduces manual intervention, and adapts to the efficient needs of automated production.

[0102] If the probe-CSP product proportion deviation value belongs to the second proportion deviation interval, the monitoring scheduling port sends a micro-compensation instruction to the lead screw driver, where the lead screw driver is a adjusting motor. The probe-CSP product proportion deviation value is processed by difference with the minimum value of the second proportion deviation interval to obtain a first deviation parameter of the probe-CSP product, and a lead screw correction compensation amount is mapped to obtain a compensation amount for the next round of angle adjustment components belonging to the lead screw.

[0103] The above-mentioned mapping of the lead screw correction compensation amount is specifically to obtain a mapping set between the first deviation parameter and the lead screw correction compensation amount from the monitoring scheduling information library, and to obtain the lead screw correction compensation amount by bringing the real-time first deviation parameter of the probe-CSP product into the mapping set.

[0104] The greater the first deviation parameter is, the more the actual deviation exceeds the minimum value of the interval, and the greater the required screw correction compensation is, so as to offset the deviation through a larger amplitude of screw displacement. Conversely, the smaller the parameter is, the smaller the compensation is. Through this mapping, the abstract deviation value can be converted into a specific screw adjustment parameter, which not only avoids the accumulation of deviation caused by insufficient compensation, but also prevents positioning shock caused by excessive compensation, thereby accurately guiding the screw action of the angle adjustment assembly, continuously optimizing the alignment accuracy of the next round, and improving the stability and efficiency of CSP product testing.

[0105] If the proportion deviation value of the probe-CSP product belongs to the third proportion deviation interval, after the alignment operation is performed, the zero point of the rotation assembly Z axis is recalibrated, wherein the Z axis represents the rotation axis of the rotation assembly around the vertical direction. The reference position (zero point) of the rotation assembly around the Z axis is recalibrated and recorded through visual means to ensure that the actual physical position of the assembly is consistent with the coordinate value in the system control. At the same time, the proportion deviation value of the probe-CSP product is processed by difference with the minimum value of the third proportion deviation interval to obtain a second deviation parameter of the probe-CSP product. The meshing gap correction factor is obtained by mapping, which is coupled with the preset meshing gap of the gear to obtain a meshing fitting gap of the gear. The meshing gap between the gear of the rotation assembly and the gear of the angle adjustment assembly is configured based on the meshing fitting gap of the gear.

[0106] The mapping obtains the meshing gap correction factor, which is specifically obtaining the mapping set between the second deviation parameter and the meshing gap correction factor from the monitoring scheduling information library. The second deviation parameter of the real-time probe-CSP product is brought into the mapping set to obtain the meshing gap correction factor.

[0107] In this embodiment, the greater the second deviation parameter is, the more the deviation exceeds the minimum value of the interval, and the greater the required meshing gap correction factor is, so as to compensate for the transmission deviation caused by mechanical wear by adjusting the gear meshing gap. The smaller the parameter is, the smaller the correction factor is, and only a slight adjustment or even maintaining the preset gap is required. Through this mapping, the abstract deviation value can be converted into a specific gear gap adjustment parameter. The meshing fitting gap obtained by coupling the preset gap can accurately configure the meshing state of the gears of the rotation assembly and the angle adjustment assembly, which not only avoids the transmission jamming (intensifies the deviation) caused by insufficient gap, but also prevents the transmission virtual position (reduces the alignment accuracy) caused by excessive gap, thereby dynamically optimizing the gear transmission accuracy and improving the accuracy of angle adjustment to ensure the alignment stability of the probe and the CSP product.

[0108] Specifically, the adsorption pressure correction is also performed on the adsorption area of the positioning assembly, and the method further comprises:

[0109] If the initial adsorption pressure deviation of the first ranking is negative after removing the absolute value, that is, the initial adsorption pressure of the adsorption head is greater than the average adsorption pressure of the adsorption area of the positioning assembly, the adsorption head is recorded as an overpressure adsorption head.

[0110] The monitoring scheduling port sends an opening reduction instruction to the electromagnetic valve of the positioning assembly, maps the initial adsorption pressure deviation of the overpressure adsorption head to obtain an adapted venting amount of the electromagnetic valve, performs difference processing on the real-time venting amount of the electromagnetic valve to obtain a venting amount deviation of the electromagnetic valve, matches to obtain an opening reduction amount of the electromagnetic valve, and configures the electromagnetic valve in real time based on the opening reduction amount of the electromagnetic valve.

[0111] The matching of the opening reduction amount of the electromagnetic valve is specifically matching the venting amount deviation of the electromagnetic valve with the opening reduction amount of the electromagnetic valve corresponding to each venting amount deviation interval predefined, determining the interval to which the venting amount deviation of the electromagnetic valve belongs, and obtaining the opening reduction amount of the electromagnetic valve corresponding to the interval.

[0112] The current opening of the electromagnetic valve is maintained, and the adsorption pressure change rate of the overpressure adsorption head is continuously monitored and compared with a predefined adsorption pressure change rate permission interval. If the adsorption pressure change rate of the overpressure adsorption head belongs to the adsorption pressure change rate permission interval, the adsorption pressure correction is completed. If the adsorption pressure change rate of the overpressure adsorption head does not belong to the adsorption pressure change rate permission interval, it is determined that the electromagnetic valve is stuck, and a mechanical maintenance alarm is triggered.

[0113] In this embodiment, considering the overpressure adsorption head whose initial adsorption pressure deviation is negative after removing the absolute value, on the one hand, local overpressure caused by electromagnetic valve sticking, pipeline blockage, etc. can be found in time, and continuous overpressure can be avoided to cause wafer local deformation, edge damage or excessive adhesion of the adsorption head and the wafer (affecting subsequent release). On the other hand, by monitoring the overpressure state, the pressure regulation accuracy of the adsorption system can be checked reversely, which provides a basis for optimizing the pressure distribution. For example, when the pressure of other adsorption heads is insufficient, the redundant adsorption force of the overpressure adsorption head can be temporarily used to maintain the overall fixing stability, and by correcting the overpressure parameters, the pressure uniformity of the entire adsorption area is ensured, and finally the attitude stability of the wafer in the test process is ensured, and the risk of alignment deviation caused by uneven stress is reduced.

[0114] The above content is only an example and description of the structure of the present application. Those skilled in the art can make various modifications or supplements to the described specific embodiments or replace them with similar ways without deviating from the structure of the present application or exceeding the scope defined by the present application.

Claims

1. An online monitoring system for CSP product CP test loading fixture based on Internet of Things, characterized in that, Comprise: The adsorption area pre-adjustment module is used to record the area of the adsorption head combination of a plurality of positioning assemblies under the CSP product coverage as the positioning assembly adsorption area, determine the adsorption state of the positioning assembly adsorption area, upload the adsorption state to the monitoring and scheduling port based on the Internet of Things, and pre-adjust the adsorption pressure of the positioning assembly adsorption area; The CSP product coordinate determination module is used to perform CSP product adsorption based on the pre-adjusted positioning assembly adsorption area, control the center point of the CSP product to enter the visual equipment field of view by the positioning assembly, and record the initial pixel coordinates of the center point of the CSP product by the visual equipment; The reference probe calibration module is used to obtain the pixel coordinates of the reference probe, and drive the positioning assembly to move to the pixel coordinates of the reference probe through the motion assembly, and determine the alignment state of the reference probe and the center point of the CSP product; The loading jig prediction adjustment module is used to adaptively predict and adjust the components of the loading jig according to the alignment state; The specific determination process of the adsorption state of the positioning assembly adsorption area is as follows: Obtain the initial adsorption pressure of a plurality of positioning assembly adsorption heads under the positioning assembly adsorption area, record it as the initial adsorption pressure of each adsorption head, and perform variance processing on the initial adsorption pressure of each adsorption head to obtain the adsorption pressure uniformity of the positioning assembly adsorption area; Compare the adsorption pressure uniformity of the positioning assembly adsorption area with the pre-defined adsorption pressure uniformity threshold to obtain the adsorption pressure comparison result, which includes the adsorption pressure comparison first result and the adsorption pressure comparison second result; The adsorption pressure comparison first result is that the adsorption pressure uniformity of the positioning assembly adsorption area is less than or equal to the adsorption pressure uniformity threshold, and the adsorption pressure comparison second result is that the adsorption pressure uniformity of the positioning assembly adsorption area is greater than the adsorption pressure uniformity threshold; If the adsorption pressure comparison result shows the adsorption pressure comparison first result, it is determined that the adsorption state of the positioning assembly adsorption area is uniform adsorption state, and the adsorption pressure of the positioning assembly adsorption area does not need to be pre-adjusted. The positioning assembly adsorption area performs CSP product adsorption; If the adsorption pressure comparison result shows the adsorption pressure comparison second result, it is determined that the adsorption state of the positioning assembly adsorption area is non-uniform adsorption state, and the adsorption pressure correction of the positioning assembly adsorption area is performed; After the adsorption pressure correction is completed, the adsorption state of the positioning assembly adsorption area is re-evaluated to determine whether iteration correction is needed; The specific correction process of the adsorption pressure correction of the positioning assembly adsorption area is as follows: Extract the adsorption pressure mean value of the positioning assembly adsorption area, perform absolute difference processing on the adsorption pressure mean value of the positioning assembly adsorption area and the initial adsorption pressure of each adsorption head respectively to obtain the initial adsorption pressure deviation of each adsorption head, and arrange them in descending order to obtain the initial adsorption pressure deviation sequence of each adsorption head. The adsorption head corresponding to the first arrangement order is recorded as the abnormal adsorption head; The pre-adjustment running state information of the abnormal adsorption head is acquired, the running effective parameter of the abnormal adsorption head is evaluated, the running effective parameter of the abnormal adsorption head is compared with a pre-defined running effective parameter threshold, if the running effective parameter of the abnormal adsorption head is greater than or equal to the running effective parameter threshold, the monitoring scheduling port sends an adsorption pressure adjustment instruction to the positioning assembly to compensate for the initial adsorption pressure deviation of the abnormal adsorption head; If the running effective parameter of the abnormal adsorption head is less than the running effective parameter threshold, the abnormal adsorption head is isolated, and a compensation adsorption group is constructed to compensate for the missing adsorption pressure of the abnormal adsorption head.

2. The online monitoring system for CSP product CP test loading fixture based on Internet of Things according to claim 1, characterized in that: The specific construction process of the compensation adsorption group for compensating for the missing adsorption pressure of the abnormal adsorption head is as follows: The straight-line distance from the abnormal adsorption head to the center point of the adsorption area of the positioning assembly is acquired, denoted as the center distribution distance of the abnormal adsorption head, and is matched with a pre-defined center distribution distance interval, if the center distribution distance of the abnormal adsorption head belongs to the center distribution distance interval, it is determined that the abnormal adsorption head belongs to the center area, otherwise it is determined that the abnormal adsorption head belongs to the edge area; If the abnormal adsorption head belongs to the center area, a plurality of normal adsorption heads in the center area are acquired, the average adsorption pressure of the adsorption area of the positioning assembly is evenly distributed to the plurality of normal adsorption heads in the center area, and a compensation adsorption group is formed to compensate for the missing adsorption pressure of the abnormal adsorption head; If the abnormal adsorption head belongs to the edge area, the center distribution distance of the abnormal adsorption head is subtracted from the maximum value of the center distribution distance interval to obtain the center deviation distance of the abnormal adsorption head, and the reference number of adsorption heads of the compensation adsorption group is mapped, according to the adsorption pressure uniformity of the adsorption area of the positioning assembly, the number of adsorption heads is corrected, and the reference number of adsorption heads of the compensation adsorption group is coupled to obtain the adaptive number of adsorption heads of the compensation adsorption group. The specific construction process of the compensation adsorption group in the edge area is as follows: a plurality of adjacent adsorption heads and the straight-line distance from the abnormal adsorption head are arranged from small to large, and the compensation adsorption group is configured in order according to the straight-line distance arrangement order until the number of adjacent adsorption heads selected meets the adaptive number of adsorption heads. 3.The online monitoring system for CSP product CP test loading fixture based on Internet of Things according to claim 1, characterized in that: The specific evaluation process of the running effective parameter of the abnormal adsorption head is as follows: The pre-adjustment running state information of the abnormal adsorption head includes the average pressure response time of the abnormal adsorption head, the average electromagnetic valve power consumption of the abnormal adsorption head, and the surface temperature rise of the abnormal adsorption head in the pre-adjustment period; The average pressure response time limit, the average electromagnetic valve power consumption limit, and the surface temperature limit are acquired from the monitoring scheduling information library; The deviation processing results include the deviation processing result between the average pressure response time of the abnormal adsorption head and the average pressure response time limit, the deviation processing result between the average electromagnetic valve power consumption of the abnormal adsorption head and the average electromagnetic valve power consumption limit, and the deviation processing result between the surface temperature rise of the abnormal adsorption head and the surface temperature limit. The influence factor and the deviation processing result are sequentially weighted and aggregated to obtain the running effective parameter of the abnormal adsorption head.

4. The online monitoring system for CSP product CP test loading fixture based on Internet of Things according to claim 1, characterized in that: The specific movement process of the motion assembly driving the positioning assembly to move to the pixel coordinates of the reference probe is as follows: According to the pixel coordinates of the reference probe, a reference probe pixel coordinate system is constructed, and according to the initial pixel coordinates of the center point of the CSP product, a CSP product pixel coordinate system is constructed; An angle between the X axis of the CSP product pixel coordinate system and the X axis of the reference probe pixel coordinate system is denoted as a rotation angle, an X axis direction offset of the origin of the CSP product pixel coordinate system relative to the origin of the reference probe pixel coordinate system is denoted as an X translation amount, and a Y axis direction offset of the origin of the CSP product pixel coordinate system relative to the origin of the reference probe pixel coordinate system is denoted as a Y translation amount; A mapping matrix is formed according to the rotation angle, the X translation amount, and the Y translation amount, and a positioning assembly is driven to move to the pixel coordinates of the reference probe by a motion assembly.

5. The online monitoring system for CSP product CP test loading fixture based on Internet of Things according to claim 1, characterized in that: The alignment state of the reference probe and the center point of the CSP product is determined, and the specific analysis process is as follows: The motion assembly drives the positioning assembly to move to the pixel coordinates of the reference probe, and the final pixel coordinates of the center point of the CSP product are obtained, and the pixel coordinates of the reference probe are calibrated, and the specific calibration process is as follows: The final pixel coordinates of the center point of the CSP product are subjected to vector processing with the pixel coordinates of the reference probe to obtain an alignment deviation of the probe-CSP product; According to the uniformity of the adsorption pressure of the adsorption region of the positioning assembly, an alignment deviation adjustment factor is obtained by mapping, and the alignment deviation adjustment factor is coupled with a preset alignment deviation to obtain an alignment defined deviation; The alignment deviation of the probe-CSP product is compared with the alignment defined deviation to obtain an alignment deviation comparison result, and the alignment deviation comparison result includes an alignment deviation first comparison result and an alignment deviation second comparison result; The alignment deviation first comparison result indicates that the alignment deviation of the probe-CSP product is less than or equal to the alignment defined deviation, and the alignment deviation second comparison result indicates that the alignment deviation of the probe-CSP product is greater than the alignment defined deviation; If the alignment deviation comparison result shows the alignment deviation first comparison result, it is determined that the alignment state of the probe-CSP product is calibration matching, and mechanical adjustment is not required, and if the alignment deviation comparison result shows the alignment deviation second comparison result, it is determined that the alignment state of the probe-CSP product is calibration mismatching, and mechanical adjustment is performed.

6. The online monitoring system for CSP product CP test loading fixture based on Internet of Things according to claim 5, characterized in that: The specific analysis process of performing mechanical adjustment is as follows: The data processing unit of the visual equipment projects the final pixel coordinates of the center point of the CSP product in the reference probe pixel coordinate system along the X axis, and the projection value of the X axis of the CSP product center point in the reference probe pixel coordinate system is denoted as an X compensation translation amount of the CSP product center point, and the projection value of the Y axis of the CSP product center point in the reference probe pixel coordinate system is denoted as a Y compensation translation amount of the CSP product center point; The data processing unit of the visual equipment projects the final pixel coordinates of the center point of the CSP product in the reference probe pixel coordinate system along the X axis, and the projection value of the X axis of the CSP product center point in the reference probe pixel coordinate system is denoted as an X compensation translation amount of the CSP product center point, and the projection value of the Y axis of the CSP product center point in the reference probe pixel coordinate system is denoted as a Y compensation translation amount of the CSP product center point; A compensation mapping matrix is formed according to the compensation rotation angle, the X compensation translation amount and the Y compensation translation amount, and the compensation mapping matrix is uploaded to the motion assembly based on the Internet of Things; the monitoring scheduling port generates a mechanical adjustment instruction; the motion assembly receives the mechanical adjustment instruction and drives the positioning assembly to move again towards the pixel coordinates of the reference probe according to the compensation mapping matrix; The rotation angle is executed by the angle adjustment assembly connected by the rotation assembly, and the X compensation translation amount and the Y compensation translation amount are executed by the motion assembly.

7. The online monitoring system for CSP product CP test loading fixture based on Internet of Things according to claim 1, characterized in that: The adjustment process is as follows: The mapping matrix and the compensation mapping matrix are respectively processed by determinant, so as to obtain the characteristic evaluation value of the mapping matrix and the characteristic evaluation value of the compensation mapping matrix respectively; the characteristic evaluation value of the compensation mapping matrix is compared with the characteristic evaluation value of the mapping matrix, and the proportion deviation value of the probe-CSP product is obtained by ratio processing; the proportion deviation value of the probe-CSP product is compared with the pre-defined proportion deviation interval, and the proportion deviation interval includes the first proportion deviation interval, the second proportion deviation interval and the third proportion deviation interval; If the proportion deviation value of the probe-CSP product belongs to the first proportion deviation interval, the proportion deviation value of the probe-CSP product is written into the monitoring scheduling port as the compensation amount of the next round of alignment operation; If the proportion deviation value of the probe-CSP product belongs to the second proportion deviation interval, the monitoring scheduling port sends a micro compensation instruction to the lead screw driver; the first deviation parameter of the probe-CSP product is obtained by difference processing of the proportion deviation value of the probe-CSP product and the minimum value of the second proportion deviation interval; the lead screw correction compensation amount is obtained by mapping, which is used as the compensation amount of the next round of angle adjustment assembly belonging to the lead screw; If the proportion deviation value of the probe-CSP product belongs to the third proportion deviation interval, after the alignment operation is executed, the zero point of the Z axis of the rotation assembly is recalibrated, and the second deviation parameter of the probe-CSP product is obtained by difference processing of the proportion deviation value of the probe-CSP product and the minimum value of the third proportion deviation interval; the meshing gap correction factor is obtained by mapping, which is coupled with the gear meshing preset gap to obtain the gear meshing adaptive gap, and the meshing gap between the rotation assembly gear and the angle adjustment assembly gear is configured based on the gear meshing adaptive gap. The method also includes:

8. The online monitoring system for CSP product CP test loading fixture based on Internet of Things according to claim 3, characterized in that: If the absolute value of the first initial adsorption pressure deviation is negative, that is, the initial adsorption pressure of the adsorption head is greater than the average adsorption pressure of the adsorption area of the positioning assembly, the adsorption head is recorded as an overpressure adsorption head; The monitoring scheduling port sends an opening reduction instruction to the electromagnetic valve of the positioning assembly; the electromagnetic valve adaptive ventilation amount is obtained by mapping according to the initial adsorption pressure deviation of the overpressure adsorption head; the electromagnetic valve ventilation amount deviation is obtained by difference processing of the electromagnetic valve real-time ventilation amount; the electromagnetic valve opening reduction amount is matched; and the electromagnetic valve is configured in real time based on the electromagnetic valve opening reduction amount. ​ Maintain the current electromagnetic valve opening, continuously monitor the adsorption pressure rate of change of the overpressure adsorption head, and compare it with the predefined adsorption pressure rate of change permission interval. If the adsorption pressure rate of change of the overpressure adsorption head belongs to the adsorption pressure rate of change permission interval, the adsorption pressure correction is completed. If the adsorption pressure rate of change of the overpressure adsorption head does not belong to the adsorption pressure rate of change permission interval, it is determined that the electromagnetic valve is stuck, and a mechanical maintenance alarm is triggered.

Images