Control system and method for full-automatic probe platform

Abstract

The invention discloses a control system and method for a full-automatic probe platform, and relates to the technical field of semiconductor manufacturing and testing equipment, and the system comprises a contact event sensing and processing module which is connected with a signal output interface of a testing machine, applies a DC bias voltage to a selected testing channel, continuously monitors a loop current at a high sampling rate, and transmits the current to the testing machine; generating an electrical feedback signal, calculating the current change rate of the electrical feedback signal in real time, and calibrating a preset threshold value based on background noise for comparison and analysis; a contact starting event is judged by capturing current in real time, the control mode is switched immediately, the system can actively respond but not passively bear the change of the contact state, intelligent and self-adaptive control of the contact process is achieved, the control mode switching adopts a direct force control mode as a main control strategy of the contact process, and the control mode switching speed is high. The overall contact pressure is ensured not to exceed a safety threshold value, and overpressure damage and poor contact are effectively prevented.

Description

Technical Field

[0001] This invention relates to the field of semiconductor manufacturing and testing equipment technology, and specifically to a control system and method for a fully automated probe platform. Background Technology

[0002] In fully automated semiconductor probe testing, the precise and reliable electrical contact between the dense probes on the probe card and the micron-sized test pads on the wafer is the physical basis for obtaining accurate test data. This contact process requires the stage (Chuck) that carries the wafer to perform precise vertical (Z-axis) movements. The ideal contact state is that all probes can simultaneously contact the corresponding test pads with uniform and moderate pressure to ensure that the contact resistance is minimized and that the wafer or probes are not damaged.

[0003] However, in actual production environments, given that wafers are not ideally flat planes, actual wafers may experience global or local warping deformation due to manufacturing stress, thermal processes, and clamping and transportation. At the same time, the probe card itself also has slight manufacturing height differences. These factors together result in the probe array and the wafer surface being in a non-ideal fit at the moment of actual contact, causing the contact sequence of each probe and the actual pressure conditions to be inconsistent. This can lead to some probes being subjected to excessive pressure due to premature contact, scratching the test pad or wearing themselves out; while other probes may have poor contact due to wafer concavity, resulting in unstable test signals or open circuits.

[0004] Faced with the aforementioned non-ideal wafer planar flatness, traditional control methods typically rely on high-precision position control, i.e., driving the stage to a preset "theoretical contact height." However, this control method completely ignores the key physical quantity of "contact force," cannot adapt to the actual shape of the wafer, and therefore cannot cope with the effects of wafer warping. The consequences are as follows: the probes that make initial contact are overloaded due to wafer tilt or bulge, easily scratching the test pads or wearing out prematurely, and may even lead to wafer fragmentation; while probes in the recessed areas may have poor contact, resulting in intermittent test failures, introducing test noise or open circuits, which seriously affect test yield and overall equipment efficiency (OEE).

[0005] Currently, how to enable the control system to sensitively sense changes in the contact state at the moment of contact and immediately and smoothly switch from a high-speed position movement mode to a precise force control mode, while also dynamically fine-tuning the worktable posture or applying compensation in zones based on real-time feedback of pressure distribution information, so as to intelligently achieve uniform and appropriate pressure throughout the probe array in complex actual contact scenarios, has become a key control problem that directly affects the reliability of testing and the safety of equipment. Therefore, there is an urgent need to provide a control system and method for a fully automated probe platform that can sense and intelligently regulate the contact pressure distribution in real time during the dynamic process of probe contact.

[0006] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0007] The purpose of this invention is to provide a fully automated probe flat tower The control system and method used in this invention solves the problems in the background art by setting up a control system with a contact event sensing and processing module, a core control mode switching module, a pressure distribution calculation and compensation instruction generation and active deformation execution module, combined with a partitioned fine-tuning structure set on a fully automatic probe platform.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a control system for a fully automatic probe platform, comprising a contact event sensing and processing module: by connecting to the signal output interface of the test machine, applying a DC bias voltage to the selected test channel, continuously monitoring the loop current at a high sampling rate, generating an electrical feedback signal, and calculating the current change rate of the electrical feedback signal in real time, and performing comparative analysis based on a preset threshold calibrated by the background noise, so as to determine the contact initiation event and output a trigger signal; Core control mode switching module: After receiving the trigger signal, it sends a control mode switching command to the main servo driver in the z-axis motion component, switching the initial cyclic position control mode CSP without disturbance to the direct force control mode DSF, and then changing the target set value of the main servo driver to the preset total contact pressure. Pressure distribution calculation and compensation command generation module: After the DSF direct force control mode is running stably, the readings of the independently distributed pressure sensor array under the worktable bearing surface are collected synchronously to form the current pressure distribution vector. The pre-stored displacement-pressure relationship influence coefficient matrix is ​​called to solve the target displacement of the fine-tuning actuator array in reverse and generate the corresponding partition compensation displacement command for driving the fine-tuning actuator array. Active deformation execution module: The partition compensation displacement command is synchronously applied to each fine-tuning actuator through digital communication to stretch or shrink to the command position, so as to actively adapt to the current back surface topography of the wafer.

[0009] Optionally, in the pre-contact stage before contact, when the workbench carrying the wafer rises to a preset position relative to the probe card, a DC bias voltage is applied to the selected test channel through the signal output interface of the test machine connected to the communication connection, and the loop current timing sequence data of the selected test channel is continuously collected using a high-speed data acquisition card to generate an electrical feedback signal; Calculation of current change rate: The current change rate is obtained by performing first-order differential calculation on the electrical feedback signal collected within a sliding window during actual operation; Calibration of preset threshold: Extract a fixed-length loop current time series data segment from the loop current time series data, calculate the statistical characteristics of the mean and standard deviation, and use the standard deviation as the evaluation value of the background noise level of the current test environment; Based on the background noise level assessment value, a dynamic judgment mechanism with dual thresholds is adopted, including a primary threshold and a secondary threshold. The current change rate calculated during actual operation is compared with a preset dual threshold to determine a valid contact initiation event; The point at which the primary threshold is first stably exceeded is recorded as the potential contact moment. Then, the time is traced back to the point at which the current signal first exceeds the secondary threshold, and this point is precisely calibrated as the final contact start time, outputting a time reference signal.

[0010] Optionally, the time point when the primary threshold is first stably exceeded is recorded as the potential contact time, and then backtracked to the time point when the current signal first exceeds the secondary threshold, and precisely calibrated as the final contact start time, and a time reference signal is output. Based on the probe type and the corresponding nominal spring force range of the probe, determine the ideal contact pressure range of a single probe on the probe card ID. The pressure loss coefficient is obtained by performing a pressure loss calibration experiment. Specifically, at room temperature (20°C), the wafer on the worktable is controlled to contact the standard calibration probe card at a standard speed. The ratio of the theoretically calculated contact pressure to the actual measured total pressure of the pressure sensor array is recorded to obtain the pressure loss coefficient. Then, the experiment is repeated at different high and low temperatures (0-60°C) to establish the temperature-loss coefficient relationship curve. The safety margin factor is dynamically calculated based on the probe material and the temperature of the test environment, combined with the test requirements. Calculate the initial theoretical total pressure based on the total number of probes, probe type, average ideal contact pressure of each probe, pressure loss coefficient, and safety margin coefficient. Using standard wafers, verification contact tests were conducted with the initial theoretical total pressure as the target. The pass rate was calculated by monitoring the contact resistance of all test channels, and the pressure uniformity index was evaluated by checking the probe indentation. The verification was repeated until all performance indicators were met, and the adjusted theoretical total pressure was obtained as the preset total contact pressure. The preset total contact pressure, probe card ID and corresponding specifications, and real-time temperature are stored in the database, so that the preset total contact pressure value can be automatically matched and obtained according to the current working conditions during the actual wafer testing stage.

[0011] Optionally, the pre-storage step of the displacement-pressure relationship influence coefficient matrix is ​​as follows: Before the experiment is calibrated, the workbench and probe card are separated to a safe distance to ensure that there is no physical contact. The system enters calibration mode, a standard flat calibration plate is placed on the workbench bearing surface, the ambient temperature is controlled at room temperature 20±0.5℃, the equipment is preheated to a thermally stable state, the actuator array and pressure sensor array are finely adjusted to perform zero-point calibration, and the initial readings of each channel are recorded. By combining optical marks on the worktable with a vision system, a four-dimensional coordinate system O-xyz-θ is established on the worktable. With the center of the worktable as the origin, the X-axis, Y-axis and θ rotation axis are on the same plane of the origin, and the Z-axis is perpendicular to the plane of the origin. Based on the pressure sensor array and the fine-tuning actuator array, the four-dimensional coordinates of each pressure sensor and each fine-tuning actuator are recorded respectively. An excitation experiment was conducted on a single actuator of the fine-tuning actuator array. By controlling one fine-tuning actuator to generate a precise standard unit displacement, the other fine-tuning actuators remained stationary, waiting for the fine-tuning actuator array to stabilize. The steady-state readings of all pressure sensors in the pressure sensor array are collected synchronously, and a fine-tuning actuator returns to the initial position to wait for recovery. Control a fine-tuning actuator to generate a reverse standard unit displacement, and wait for the fine-tuning actuator array to stabilize before synchronously acquiring the steady-state reading changes of all pressure sensors again. Calculate the average influence coefficient of each fine-tuning actuator on each pressure sensor to form an initial displacement-pressure relationship influence coefficient matrix; By combining the mechanical cross-coupling effect between fine-tuning actuators to correct the initial displacement-pressure relationship influence coefficient matrix, several different actuator displacement combinations are randomly selected to actually perform displacement and measure the pressure changes of all pressure sensors. The pressure change is predicted based on the initial displacement-pressure relationship influence coefficient matrix; The residual matrix is ​​calculated based on the predicted pressure changes, and the initial displacement-pressure relationship influence coefficient matrix is ​​corrected using the least squares method to obtain the corrected displacement-pressure relationship influence coefficient matrix. The modified displacement-pressure relationship influence coefficient matrix was verified using the new fine-tuning actuator displacement combination.

[0012] Optionally, the steps for generating the partition compensation displacement command are as follows: The raw readings of all pressure sensors are acquired synchronously, and a digital filter is applied to each channel to obtain the filtered pressure value. Construct the pressure distribution vector at the current moment, and subtract the zero-point pressure baseline pre-stored based on the non-contact state calibration to obtain the net pressure vector; The weighted average of the current pressure distribution is calculated based on the net pressure vector, taking into account the position weights of each pressure sensor, and the pressure deviation vector is calculated. The influence of the current temperature of the workbench on the pressure loss is read, and the maximum allowable displacement of the fine-tuning actuator is used as a constraint to solve the objective function to obtain the compensation displacement command vector. To avoid abrupt changes in the command, the displacement command obtained by the solution is smoothed, and a feedforward compensation term is added in combination with the system dynamic response to obtain the final partitioned compensation displacement command.

[0013] Active deformation execution module: The partition compensation displacement command is synchronously applied to each fine-tuning actuator through digital communication to stretch or shrink to the command position, so as to actively adapt to the current back surface topography of the wafer.

[0014] Optionally, the contact event sensing and processing module is communicatively connected to the signal output interface of the test machine; The core control mode switching module is communicatively connected to the main servo driver in the z-axis motion component that drives the vertical movement of the worktable. The pressure distribution calculation and compensation instruction generation module is digitally connected to both the pressure sensor array and the fine-tuning actuator array.

[0015] A control method for a fully automated probe platform includes the following steps: S1. Contact Initiation Event Determination: During the movement of the fully automated probe stage toward the probe card, the electrical feedback signal from the test machine is continuously monitored. By identifying that the current change rate of the electrical feedback signal exceeds the preset threshold calibrated based on the background noise, the contact initiation event between the probe and the wafer test pad is determined in real time. S2, Control Mode Switching: After determining the contact initiation event, the control mode of the servo axis on the main servo drive that drives the vertical movement of the worktable is immediately switched from position control mode to direct force control mode with the goal of maintaining the preset total contact pressure. S3. Pressure distribution calculation and command generation: In direct force control mode, the readings of the independent pressure sensor array arranged under the workbench bearing surface are collected synchronously. Based on the pre-calibrated relationship matrix that reflects the influence of the displacement of each fine-tuning actuator on the pressure of each pressure sensor, the deviation between each pressure reading and the preset total contact pressure is calculated into a set of partition compensation displacement commands for driving the corresponding area fine-tuning actuator. S4. Active Deformation Execution: The partition compensation displacement command is synchronously applied to each fine-tuning actuator, so that the bearing surface of the worktable generates controlled elastic deformation through the deformation of the flexible support body below it, so as to actively adapt to the back surface morphology of the current wafer and promote the uniformity of the contact pressure of each probe on the probe card.

[0016] A fully automated probe platform, and a control system for the fully automated probe platform, including an industrial control distribution box; The electrical control system is integrated inside the industrial control power distribution box, and the vision system is installed on the upper surface of the industrial control power distribution box and connected to the electrical control system. The human-machine interface, flexible positioning support frame, and material handling robot arm are movably installed on the upper surface of the industrial control power distribution box from left to right. A precision positioning four-dimensional motion module is set in the middle of the flexible positioning support frame and inside the industrial control power distribution box. One end of the material handling robot arm is movably connected to the basket loading position. A pre-alignment component is set between the material handling robot arm and the basket loading position. The precision positioning four-dimensional motion module includes a detection tray, a worktable is provided in the middle of the upper surface of the detection tray, and the sides of the worktable are provided with an x-axis motion component, a y-axis motion component, a z-axis motion component and a θ-rotation axis motion component from bottom to top, and a probe card is placed on the upper surface of the worktable.

[0017] Optionally, the workbench is provided with a partitioned fine-tuning structure, the partitioned fine-tuning structure including: a rigid bearing plate; A flexible support body is a metal structure connected to the bottom of the rigid bearing plate and equipped with flexible hinges or weak stiffeners. A fine-tuning actuator array is arranged between the flexible support and the z-axis motion component; A pressure sensor array is configured to measure the support reaction force of each partition of the partitioned fine-tuning structure.

[0018] Optionally, the fine-tuning actuator array is a piezoelectric ceramic actuator array or a voice coil motor array; The pressure sensor array is a strain gauge sensor array, integrated into the flexible support body.

[0019] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention determines the contact initiation event by capturing the current in real time and immediately switches the control mode. The system can actively respond to rather than passively withstand changes in the contact state, realizing intelligent and adaptive control of the contact process. Furthermore, the switching control mode adopts the direct force control mode as the main control strategy for the contact process, ensuring that the overall contact pressure does not exceed the safety threshold, avoiding the risk of probe or test pad crushing caused by local wafer bulging, and effectively preventing overpressure damage and poor contact. By calculating the pressure distribution and driving the worktable to actively generate adaptive deformation, wafer warpage can be dynamically compensated, and the pressure can be evenly distributed on hundreds to thousands of probes, achieving pressure uniformity, greatly improving the consistency of contact resistance, improving test accuracy and stability, and the uniform pressure distribution reduces the abnormal load on a single probe, extends the service life of the probe card, and reduces the frequency of test interruptions and maintenance caused by contact problems. This control system and method, along with a partitioned fine-tuning structure on the platform, solves the problem of uneven probe contact pressure caused by wafer warping and probe height differences. By combining real-time event sensing, intelligent control mode switching, pressure distribution calculation, and active attitude compensation, it achieves precise control and uniform distribution of the overall contact pressure of the probe array, thereby avoiding test pad damage, abnormal probe wear, and ensuring stable and reliable test electrical contact. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0021] Figure 1 This is a block diagram of the control system used in the fully automated probe platform of this invention.

[0022] Figure 2 This is a flowchart of the control method used in the fully automated probe platform of the present invention.

[0023] Figure 3 This is a three-dimensional structural diagram of the fully automated probe platform of the present invention.

[0024] Figure 4 This is a front view of the structure of the fully automated probe platform of the present invention.

[0025] Figure 5 This is a side view of the structure of the fully automated probe platform of the present invention.

[0026] Figure 6 This is a schematic diagram of the structure of the precision positioning four-dimensional motion module of the present invention.

[0027] The following are the reference numerals in the attached diagram: 1. Industrial control motor box; 2. Electrical control system; 3. Human-machine interface; 4. Precision positioning four-dimensional motion module; 401. Inspection tray; 402. X-axis motion assembly; 403. Y-axis motion assembly; 404. Z-axis motion assembly; 405. θ-axis rotation motion assembly; 406. Worktable; 407. Probe card; 5. Flexible positioning support frame; 6. Material handling robot arm; 7. Basket loading position; 8. Pre-alignment assembly; 9. Vision system. Detailed Implementation

[0028] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0029] Example 1 This invention provides, for example Figure 1 The control system for a fully automated probe platform shown includes a contact event sensing and processing module: it is activated when the workbench carrying the wafer rises to a distance of tens of micrometers from the probe card. By connecting to the signal output interface of the test machine, it applies a DC bias voltage of 0.1V to the selected test channel and continuously monitors the loop current at a high sampling rate of 100kHz, generates an electrical feedback signal, and calculates the first-order difference of the electrical feedback signal in real time to obtain the current change rate. It then compares and analyzes the results with a preset threshold calibrated based on the background noise during the debugging phase to determine the contact initiation event and outputs the trigger time reference signal to the core control mode switching module. In a preferred embodiment, the steps for determining the contact initiation event are as follows: In the pre-contact stage, when the worktable carrying the wafer rises to a preset position relative to the probe card, a 0.1V DC bias voltage is applied to the selected test channel via the communication connection tester signal output interface. A high-speed data acquisition card continuously acquires the loop current timing sequence data of the selected test channel at a 100kHz sampling rate, generating an electrical feedback signal and calibrating it. ; Calculation of current change rate: The current change rate is obtained by performing first-order differential calculation on the electrical feedback signal collected within a sliding window during actual operation. The formula for calculating the current change rate is as follows: In the formula, Expressed as the rate of change of current, Indicated as in Electrical feedback signals collected during actual operation. This is represented as a first-order difference operation of current; Calibration of preset threshold: Extract a loop current time series data segment with a fixed time length of 100ms from the loop current time series data, calculate the statistical characteristics of the mean and standard deviation, and use the standard deviation as the evaluation value of the background noise level of the current test environment; Based on the background noise level assessment, a dynamic judgment mechanism with dual thresholds is adopted, including a primary threshold and a secondary threshold. The expression for the primary threshold is as follows: ,and ,as well as In the formula, Represented as the primary threshold, Represented as empirical coefficients, The standard deviation of current is represented as the current standard deviation of the upper loop current time series data with a fixed time length. This represents the total number of samples in the upper loop current time series data with a fixed time length. Indicated as in The loop current collected at the sampling time, This represents the mean current value of the upper loop current time series data with a fixed time length. Represented as a sampling index; The expression for the secondary threshold is: In the formula, This is represented as a secondary threshold. This is expressed as the attenuation coefficient; The current change rate calculated in actual operation is compared with the preset dual thresholds, including the judgment condition: first, when the average current change rate in the sliding window exceeds the main threshold for the first time. Secondly, the rate of change of current at the sampling points within the sliding window exceeds the event for at least 10% of all consecutive sampling points; Third, within a sliding window of the same size before and after the main threshold exceedance point, the monotonicity of the current signal remains stable. When all three conditions above are met, it is determined to be a valid contact initiation event; The point at which the primary threshold is first stably exceeded is recorded as the potential contact moment. Then, the time is traced back to the point at which the current signal first exceeds the secondary threshold, and this point is precisely calibrated as the final contact start time, outputting a time reference signal.

[0030] Core control mode switching module: After receiving the trigger signal, it sends a control mode switching command to the main servo driver of the Z-axis vertical motion motor in the Z-axis motion component through the EtherCAT bus, switching the initial cyclic position control mode CSP without disturbance to the direct force control mode DSF. Then, the target set value of the main servo driver is changed to the preset total contact pressure, and the preset total contact pressure is the real-time output force signal estimated by the motor current loop. In a preferred embodiment, the preset logic for the total contact pressure is as follows: The probe card specifications are automatically retrieved by identifying the probe card ID, including the total number of probes, probe type, nominal spring force range of the probe, probe material and maximum allowable stress value. The probe types include cantilever probes, vertical probes and MEMS probes. Based on the probe type and the corresponding nominal spring force range, the ideal contact pressure range for a single probe on the probe card ID is determined. The recommended pressure for ideal contact of a single probe is... In the formula, Represented as the first Recommended single-needle pressure for this type of probe. It is expressed as the basic proportionality coefficient, and , This represents the nominal minimum spring force of the probe. This represents the nominal maximum spring force of the probe. This is expressed as the spring force range utilization coefficient, and , It is expressed as the temperature compensation coefficient, and , Represented as current time The corresponding ambient temperature, This is a reference temperature, i.e., room temperature 20℃; The ideal contact pressure range for a single probe of a cantilever needle is: ; The ideal contact pressure range for a single probe of a vertical needle is: ; The ideal contact pressure range for a single MEMS probe is: ; The pressure loss coefficient was obtained by performing a pressure loss calibration experiment. Specifically, at room temperature (20°C), the wafer on the worktable was brought into contact with a standard calibration probe card at a standard speed. The ratio of the theoretically calculated contact pressure to the actual total pressure measured by the pressure sensor array was recorded, thus obtaining the pressure loss coefficient. The experiment was then repeated at different high and low temperatures (0-60°C) to establish a temperature-loss coefficient relationship curve. The formula for calculating the pressure loss coefficient is as follows: ,and In the formula, Expressed as the pressure loss coefficient at room temperature (20℃), This represents the total pressure actually measured by the pressure sensor array. Expressed as theoretically calculated contact pressure, This is expressed as the fitted temperature-loss coefficient relationship function. , , , These are the fitting parameters determined by the least squares method. Represented as temperature; The safety margin factor is dynamically calculated based on the probe material and the ambient temperature of the test environment, combined with the test requirements. The formula for calculating the safety margin factor is as follows: ,and In the formula, This is expressed as a safety margin coefficient. This represents the basic safety factor of the probe material. This is represented as a material hardness correction term. This is indicated as a test duration adjustment item. This is represented as a temperature deviation correction term. This represents the weighting coefficient for determining the hardness of the corresponding material in the experimental calibration. This represents the hardness of the test pad material. Indicated as the hardness of the reference material. This represents the weighting coefficient corresponding to the test duration in the experimental calibration. This represents the estimated duration of a single-point test. This is a reference test duration, and is 0.1 seconds. This represents the weighting coefficient for the temperature deviation corresponding to the experimental calibration. This indicates the optimal test temperature is 25℃. Based on the total number of probes, probe type, average ideal contact pressure of each probe, pressure loss coefficient, and safety margin coefficient, the initial theoretical total pressure is calculated. The formula for calculating the initial theoretical total pressure is as follows: In the formula, Represented as the first The initial theoretical total pressure of the type probe, Represented as the first The total number of probes of each type. This is expressed as the average recommended pressure per needle; Using standard wafers, verification contact tests were performed with the initial theoretical total pressure as the target. The pass rate was calculated by monitoring the contact resistance of all test channels, and the pressure uniformity was evaluated by checking probe indentations. When the pass rate was below 99.9% or the pressure uniformity exceeded 30%, the adjustment strategy included increasing the initial theoretical total pressure by 5%-10% if an open circuit was detected by probe indentation, and decreasing the initial theoretical total pressure by 5%-10% if the indentation was too deep. This verification was repeated until all performance indicators were met. The adjusted theoretical total pressure was then used as the preset total contact pressure. The formula for calculating the preset total contact pressure is as follows: ,and In the formula, This is expressed as the preset total contact pressure. This is represented as the probe aging correction factor. Expressed as the aging degradation rate, and , This represents the cumulative usage time of the probe card. This is expressed as the nominal lifespan of the probe card; The expression for performance evaluation in the verification phase is: In the formula, This is expressed as the contact pass rate. This represents the number of channels with acceptable contact resistance. This represents the total number of test channels. This is expressed as pressure uniformity. This is expressed as the standard deviation of the pressure readings for each zone. This is expressed as the average pressure reading for each zone. This represents the estimated indentation depth. It is expressed as the material indentation coefficient. This represents the actual pressure per needle. This represents the contact area of ​​the probe tip. This is expressed as the duration of contact; The preset total contact pressure, probe card ID and corresponding specifications, and real-time temperature are stored in the database, so that the preset total contact pressure value can be automatically matched and obtained according to the current working conditions during the actual wafer testing stage.

[0031] Pressure distribution calculation and compensation command generation module: After the direct force control mode (DSF) is running stably, the readings of the independently distributed pressure sensor array under the worktable bearing surface are synchronously collected at a frequency of 1kHz, which constitutes the current pressure distribution vector. By calculating the average pressure and each pressure deviation value, the pre-stored displacement-pressure relationship influence coefficient matrix is ​​obtained by calling the experimental calibration. The target displacement of the fine-tuning actuator array that can make each current pressure deviation tend to zero is solved in reverse. The corresponding partition compensation displacement command for driving the fine-tuning actuator array is generated and transmitted to the active deformation execution module. In a preferred embodiment, the pre-storage step of the displacement-pressure relationship influence coefficient matrix is ​​as follows: Before the experiment is calibrated, the workbench and the probe card are separated to a safe distance of 30μm to ensure that there is no physical contact. The system enters calibration mode. A standard plane calibration plate is placed on the workbench bearing surface, and the flatness of the standard plane calibration plate is better than 1μm. The ambient temperature is controlled at room temperature 20±0.5℃. The equipment is preheated for 1 hour to reach a thermally stable state. The actuator array and pressure sensor array are finely adjusted for zero-point calibration, and the initial readings of each channel are recorded. By combining optical marks on the worktable with a vision system, a four-dimensional coordinate system O-xyz-θ is established on the worktable. With the center of the worktable as the origin, the X-axis, Y-axis and θ rotation axis are on the same plane of the origin, and the Z-axis is perpendicular to the plane of the origin. Based on the pressure sensor array and the fine-tuning actuator array, the four-dimensional coordinates of each pressure sensor and each fine-tuning actuator are recorded respectively. An excitation experiment was conducted on a single actuator of the fine-tuning actuator array. By controlling one fine-tuning actuator to generate a precise standard unit displacement, the other fine-tuning actuators remained stationary, waiting for the fine-tuning actuator array to stabilize, in order to ensure the damping of mechanical vibration. The steady-state readings of all pressure sensors in the pressure sensor array are collected synchronously, and a fine-tuning actuator returns to the initial position to wait for recovery. Control a fine-tuning actuator to generate a reverse standard unit displacement, and wait for the fine-tuning actuator array to stabilize before synchronously acquiring the steady-state reading changes of all pressure sensors again. Calculate the average influence coefficient of each fine-tuning actuator on each pressure sensor to construct the initial displacement-pressure relationship influence coefficient matrix. The expression for the initial influence coefficient of each fine-tuning actuator on each pressure sensor is as follows: In the formula, Represented as the first The fine-tuning actuator is for the first The initial influence coefficient of each pressure sensor, Represented as the first in the fine-tuning actuator array A fine-tuning actuator, Represented as the first in the pressure sensor array One pressure sensor, This represents the total number of repeated experiments. This is expressed as the standard unit displacement of the fine-tuning actuator. Represented as the first The first positive incentive Pressure change of each pressure sensor Represented as the first The first negative incentive Pressure change measured by each pressure sensor; By combining the mechanical cross-coupling effect between fine-tuning actuators to correct the initial displacement-pressure relationship influence coefficient matrix, several different actuator displacement combinations are randomly selected to actually perform displacement and measure the pressure changes of all pressure sensors. The pressure change is predicted based on the initial displacement-pressure relationship influence coefficient matrix; The residual matrix is ​​calculated based on the predicted pressure changes, and the initial displacement-pressure relationship influence coefficient matrix is ​​corrected using the least squares method to obtain the corrected displacement-pressure relationship influence coefficient matrix. The modified displacement-pressure relationship influence coefficient matrix was verified using a new fine-tuning actuator displacement combination. The determinism of the experimentally calibrated modified displacement-pressure relationship influence coefficient matrix was evaluated by calculating the average prediction error, maximum prediction error, and coefficient of determination.

[0032] In a preferred embodiment, the steps for generating the partition compensation displacement command are as follows: Raw readings from all pressure sensors were synchronously acquired at a frequency of 1 kHz. A digital filter was applied to each channel using a moving average filter to remove high-frequency noise, followed by a zero-phase low-pass filter to retain valid signals. Outliers were then removed based on the 3σ criterion to obtain the filtered pressure value, which was calibrated as follows: ; Construct the pressure distribution vector at the current moment, and subtract the zero-point pressure baseline pre-stored based on the non-contact state calibration to obtain the net pressure vector. The formula for calculating the net pressure vector is as follows: In the formula, Represented as a net pressure vector, This is represented as the filtered pressure value. Represented as zero-point pressure baseline; The weighted average of the current pressure distribution is calculated based on the net pressure vector, taking into account the position weights of each pressure sensor, and the pressure deviation vector is also calculated. The formula for calculating the weighted average is as follows: In the formula, This is expressed as the weighted average of the current pressure distribution. This represents the number of all pressure sensors in the pressure sensor array. Represented as the first The position weight of each pressure sensor Represented as the first The net pressure vector of each pressure sensor; The formula for calculating the pressure deviation vector is as follows: In the formula, Represented as the first Pressure deviation vector of each pressure sensor Represented as the first The weighted average of the pressure sensors; The influence of the current temperature of the workbench on the pressure loss is read, and the maximum allowable displacement of the fine-tuning actuator is used as a constraint to solve the objective function to obtain the compensation displacement command vector. To avoid abrupt changes in the command, the displacement command obtained by the solution is smoothed, and a feedforward compensation term is added in combination with the system dynamic response to obtain the final partitioned compensation displacement command.

[0033] Active Deformation Execution Module: The partition compensation displacement command is synchronously applied to each fine-tuning actuator through digital communication to extend or retract to the command position. This causes the bearing surface of the worktable to generate controlled elastic bending deformation through the deformation of the flexible support below, so as to actively adapt to the current back surface morphology of the wafer. This promotes the homogenization of the contact pressure of each probe on the probe card. After the pressure is equalized, the pressure homogenization state is maintained to perform stable electrical testing. After the test is completed, the control process is reversed, and the worktable descends smoothly to separate the wafer from the probe.

[0034] In a preferred embodiment, the contact event sensing and processing module is communicatively connected to the signal output interface of the test machine; The core control mode switching module is communicatively connected to the main servo driver in the z-axis motion component that drives the vertical movement of the worktable. The pressure distribution calculation and compensation instruction generation module is digitally connected to both the pressure sensor array and the fine-tuning actuator array.

[0035] The technical description and effects of this embodiment: Through a collaborative architecture of various modules, the system utilizes a contact event sensing module to convert physical contact events into precise time reference signals through high-frequency sampling and intelligent signal processing; the core control mode switching module uses this reference to achieve a smooth and seamless switch from position control to force control with safe pressure; the pressure distribution calculation module calculates pressure deviations in real time using a pre-stored mechanical model and generates compensation commands; the active deformation execution module drives the fine-tuning mechanism to complete precise deformation; fundamentally solving the problem of uneven pressure caused by wafer warpage, significantly improving the consistency and reliability of probe testing, while greatly reducing the risk of probe wear and wafer damage.

[0036] Example 2 This invention provides, for example Figure 2 The control method for a fully automated probe platform shown includes the following steps: S1. Contact Initiation Event Determination: During the movement of the fully automated probe stage toward the probe card, the electrical feedback signal from the test machine is continuously monitored. By identifying that the current change rate of the electrical feedback signal exceeds the preset threshold calibrated based on the background noise, the contact initiation event between the probe and the wafer test pad is determined in real time. S2, Control Mode Switching: After determining the contact initiation event, the control mode of the servo axis on the main servo drive that drives the vertical movement of the worktable is immediately switched from position control mode to direct force control mode with the goal of maintaining the preset total contact pressure. S3. Pressure distribution calculation and command generation: In direct force control mode, the readings of the independent pressure sensor array arranged under the workbench bearing surface are collected synchronously. Based on the pre-calibrated relationship matrix that reflects the influence of the displacement of each fine-tuning actuator on the pressure of each pressure sensor, the deviation between each pressure reading and the preset total contact pressure is calculated into a set of partition compensation displacement commands for driving the corresponding area fine-tuning actuator. S4. Active Deformation Execution: The partition compensation displacement command is synchronously applied to each fine-tuning actuator, so that the bearing surface of the worktable generates controlled elastic deformation through the deformation of the flexible support body below it, so as to actively adapt to the back surface morphology of the current wafer and promote the uniformity of the contact pressure of each probe on the probe card.

[0037] In a preferred embodiment, the calibration steps for the preset threshold are as follows: Separate the stage and probe card to a safe distance, typically tens of micrometers, to ensure no physical contact. The system enters calibration mode, and the same 0.1V DC bias voltage as the formal test is applied to the output of the tester, but the current measurement range is switched to the highest sensitivity setting. Collect multi-dimensional background noise, including time-dimensional noise: continuously collect multiple segments of current data at a fixed location, each segment lasting 1 second, with a sampling rate of 100kHz; Spatial dimension noise: Control the stage to perform low-speed scanning within a range of ±10μm in the O-XY horizontal plane, while simultaneously collecting current data to assess the noise introduced by mechanical motion; Frequency domain noise: Perform FFT transformation on the acquired time domain data to analyze the main noise frequency components; For all the multi-dimensional background noise data segments collected, the standard deviation is calculated and outlier data segments are removed. Initiate the dynamic contact dual threshold calibration test: control the workbench to approach the probe card at a constant speed of 0.5μm / s, and at the same time acquire the current signal at high speed. When a significant current step is detected for the first time, record this moment as the reference contact point and stop the movement immediately. Noise data in the last 100ms before contact and current signal data in the last 50ms after contact were extracted from the recorded current data. Based on calibration experimental data, the threshold parameters for two noise fluctuation limits and contact characteristic intensity were calculated and calibrated. Using the calibrated threshold, repeat the contact experiment 3-5 times under the same conditions and record the detection results each time. The evaluation indicators include detection delay time, false positive rate, and false negative rate. If any indicator fails to meet the standard, adjust the primary threshold coefficient and secondary threshold until the performance requirements are met. The performance indicators are: delay > 1ms and false positive / false negative rate > 1%.

[0038] In a preferred embodiment, the direct force control mode employs a proportional-integral controller and executes a seamless switching logic, using the torque value output by the main servo drive at the moment of mode switching as the initial value of the integral term of the force control loop.

[0039] The technical description and effects of this embodiment are as follows: This control method first monitors the rate of change of electrical signals in real time to accurately determine the initial contact moment; then it immediately switches the Z-axis control mode to the force control mode to establish a total pressure safety boundary; then, based on the distribution data of the pressure sensor array, it uses a pre-calibrated influence coefficient matrix to inversely solve for the compensation displacement; finally, it drives the fine-tuning actuator array to cause adaptive deformation of the bearing surface; further, it ensures system safety through event-triggered control mode switching, achieves accurate compensation through spatial calculation of the mathematical model, and achieves physical adaptation through mechanical deformation execution. This achieves microscopic homogenization of the contact pressure of hundreds to thousands of probes, eliminates test data fluctuations caused by uneven pressure, significantly improves test yield, and effectively ensures probe life and wafer integrity.

[0040] Example 3 This invention provides, for example Figure 3-6 The fully automatic probe platform shown includes an industrial control power distribution box 1, and also includes a control system for using a fully automatic probe platform, wherein the industrial control power distribution box 1; The electrical control system 2 is integrated inside the industrial control power distribution box 1. It is used to power and drive the precision positioning four-dimensional motion module 4, the material handling robot arm 6, the pre-alignment component 8 and the vision system 9. The vision system 9 is located on the upper surface of the industrial control power distribution box 1 and is connected to the electrical control system 2. It is used to collect the contact images between the wafer and the probe. The human-machine interface 3, the flexible positioning support frame 5, and the picking and placing robotic arm 6 are movably installed on the upper surface of the industrial control motor box 1 from left to right. The fine positioning four-dimensional motion module 4 is set in the middle of the flexible positioning support frame 5 and inside the industrial control motor box 1. One end of the picking and placing robotic arm 6 is movably connected to the basket loading position 7. A pre-alignment component 8 is set between the picking and placing robotic arm 6 and the basket loading position 7. The precision positioning four-dimensional motion module 4 includes a detection tray 401. A worktable 406 is provided in the middle of the upper surface of the detection tray 401. From bottom to top, the sides of the worktable 406 are provided with an x-axis motion component 402, a y-axis motion component 403, a z-axis motion component 404 and a θ-axis rotation motion component 405. A probe card 407 is placed on the upper surface of the worktable 406.

[0041] In a preferred embodiment, the worktable 406 is provided with a partitioned fine-tuning structure, which includes a rigid bearing plate; A flexible support body, a metal structure connected to the bottom of a rigid bearing plate, and equipped with flexible hinges or weak stiffeners; A fine-tuning actuator array is arranged between the flexible support and the z-axis motion component 404, and is connected to the active deformation execution module of the control system. A pressure sensor array is configured to measure the support reaction force of each zone in the partitioned fine-tuning structure and is connected to the pressure distribution calculation and compensation command generation module of the control system.

[0042] In a preferred embodiment, the fine-tuning actuator array is a piezoelectric ceramic actuator array or a voice coil motor array; The pressure sensor array is a strain gauge sensor array, integrated into a flexible support body.

[0043] The technical description and effects of this embodiment: The platform integrates a layered pressure control mechanism based on the precision positioning four-dimensional motion module 4. By setting up a zoned fine-tuning structure, under the premise of controlled total pressure, it drives the actuators in the corresponding areas to produce micron-level displacement by sensing local pressure differences. Utilizing the elastic deformation characteristics of the flexible support, the bearing plate surface produces micro-deformation that matches the wafer warping, further providing a physical basis for pressure uniformity from the mechanical structure level. This allows the compensation commands generated by the control system to be accurately translated into spatial morphological changes of the bearing plate, ultimately achieving adaptive bonding between the probe array and the wafer surface, creating an ideal mechanical environment for stable and reliable electrical contact testing.

[0044] The present invention provides a control method for a fully automated probe platform, which is implemented by the aforementioned control system for a fully automated probe platform. The control system for a fully automated probe platform is applied to a fully automated probe platform, which will not be described in detail here.

[0045] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0046] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0047] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0048] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0049] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A control system for a fully automated probe platform, characterized in that, It includes a contact event sensing and processing module: by connecting to the signal output interface of the test machine, it applies a DC bias voltage to the selected test channel and continuously monitors the loop current at a high sampling rate, generates an electrical feedback signal, calculates the current change rate of the electrical feedback signal in real time, and performs comparative analysis based on the background noise to determine the contact initiation event and output a trigger signal; Core control mode switching module: After receiving the trigger signal, it sends a control mode switching command to the main servo driver in the z-axis motion component, switching the initial cyclic position control mode CSP without disturbance to the direct force control mode DSF, and then changing the target set value of the main servo driver to the preset total contact pressure. Pressure distribution calculation and compensation command generation module: After the direct force control mode (DSF) is running stably, the readings of the independently distributed pressure sensor array under the worktable bearing surface are collected synchronously to form the current pressure distribution vector. The pre-stored displacement-pressure relationship influence coefficient matrix is ​​called to solve the target displacement of the fine-tuning actuator array in reverse and generate the corresponding partition compensation displacement command for driving the fine-tuning actuator array. Active deformation execution module: The partition compensation displacement command is synchronously applied to each fine-tuning actuator through digital communication to stretch or shrink to the command position, so as to actively adapt to the current back surface topography of the wafer.

2. The control system for a fully automated probe platform according to claim 1, characterized in that, In the pre-contact stage before contact, when the workbench carrying the wafer rises to a preset position relative to the probe card, a DC bias voltage is applied to the selected test channel through the signal output interface of the test machine via communication connection. The high-speed data acquisition card continuously acquires the loop current timing sequence data of the selected test channel to generate an electrical feedback signal. Calculation of current change rate: The current change rate is obtained by performing first-order differential calculation on the electrical feedback signal collected within a sliding window during actual operation; Calibration of preset threshold: Extract a fixed-length loop current time series data segment from the loop current time series data, calculate the statistical characteristics of the mean and standard deviation, and use the standard deviation as the evaluation value of the background noise level of the current test environment; Based on the background noise level assessment value, a dynamic judgment mechanism with dual thresholds is adopted, including a primary threshold and a secondary threshold. The current change rate calculated during actual operation is compared with a preset dual threshold to determine a valid contact initiation event; The point at which the primary threshold is first stably exceeded is recorded as the potential contact moment. Then, the time is traced back to the point at which the current signal first exceeds the secondary threshold, and this point is precisely calibrated as the final contact start time, outputting a time reference signal.

3. The control system for a fully automated probe platform according to claim 2, characterized in that, The time point when the primary threshold is first stably exceeded is recorded as the potential contact time. Then, the time point when the current signal first exceeds the secondary threshold is traced back is accurately calibrated as the final contact start time, and a time reference signal is output. Based on the probe type and the corresponding nominal spring force range of the probe, determine the ideal contact pressure range of a single probe on the probe card ID. The pressure loss coefficient is obtained by performing a pressure loss calibration experiment. That is, at room temperature of 20℃, the wafer on the worktable is controlled to contact the standard calibration probe card at a standard speed, and the ratio of the theoretically calculated contact pressure to the actual measured total pressure of the pressure sensor array is recorded to obtain the pressure loss coefficient. Then, the experiment is repeated at different temperatures different from room temperature to establish the temperature-loss coefficient relationship curve. The safety margin factor is dynamically calculated based on the probe material and the temperature of the test environment, combined with the test requirements. Calculate the initial theoretical total pressure based on the total number of probes, probe type, average ideal contact pressure of each probe, pressure loss coefficient, and safety margin coefficient. Using standard wafers, verification contact tests were conducted with the initial theoretical total pressure as the target. The pass rate was calculated by monitoring the contact resistance of all test channels, and the pressure uniformity index was evaluated by checking the probe indentation. The verification was repeated until all performance indicators were met, and the adjusted theoretical total pressure was obtained as the preset total contact pressure. The preset total contact pressure, probe card ID and corresponding specifications, and real-time temperature are stored in the database, so that the preset total contact pressure value can be automatically matched and obtained according to the current working conditions during the actual wafer testing stage.

4. The control system for a fully automated probe platform according to claim 3, characterized in that, The steps for pre-storing the displacement-pressure relationship influence coefficient matrix are as follows: Before the experiment is calibrated, the workbench and probe card are separated to a safe distance to ensure that there is no physical contact. The system enters calibration mode, a standard flat calibration plate is placed on the workbench bearing surface, the ambient temperature is controlled at room temperature 20±0.5℃, the equipment is preheated to a thermally stable state, the actuator array and pressure sensor array are finely adjusted to perform zero-point calibration, and the initial readings of each channel are recorded. By combining optical marks on the worktable with a vision system, a four-dimensional coordinate system O-xyz-θ is established on the worktable. With the center of the worktable as the origin, the X-axis, Y-axis and θ rotation axis are on the same plane of the origin, and the Z-axis is perpendicular to the plane of the origin. Based on the pressure sensor array and the fine-tuning actuator array, the four-dimensional coordinates of each pressure sensor and each fine-tuning actuator are recorded respectively. An excitation experiment was conducted on a single actuator of the fine-tuning actuator array. By controlling one fine-tuning actuator to generate a precise standard unit displacement, the other fine-tuning actuators remained stationary, waiting for the fine-tuning actuator array to stabilize. The steady-state readings of all pressure sensors in the pressure sensor array are collected synchronously, and a fine-tuning actuator returns to the initial position to wait for recovery. Control a fine-tuning actuator to generate a reverse standard unit displacement, and wait for the fine-tuning actuator array to stabilize before synchronously acquiring the steady-state reading changes of all pressure sensors again. Calculate the average influence coefficient of each fine-tuning actuator on each pressure sensor to form an initial displacement-pressure relationship influence coefficient matrix; By combining the mechanical cross-coupling effect between fine-tuning actuators to correct the initial displacement-pressure relationship influence coefficient matrix, several different actuator displacement combinations are randomly selected to actually perform displacement and measure the pressure changes of all pressure sensors. The pressure change is predicted based on the initial displacement-pressure relationship influence coefficient matrix; The residual matrix is ​​calculated based on the predicted pressure changes, and the initial displacement-pressure relationship influence coefficient matrix is ​​corrected using the least squares method to obtain the corrected displacement-pressure relationship influence coefficient matrix. The modified displacement-pressure relationship influence coefficient matrix was verified using the new fine-tuning actuator displacement combination.

5. The control system for a fully automated probe platform according to claim 4, characterized in that, The steps for generating the partition compensation displacement command are as follows: The raw readings of all pressure sensors are acquired synchronously, and a digital filter is applied to each channel to obtain the filtered pressure value. Construct the pressure distribution vector at the current moment, and subtract the zero-point pressure baseline pre-stored based on the non-contact state calibration to obtain the net pressure vector; The weighted average of the current pressure distribution is calculated based on the net pressure vector, taking into account the position weights of each pressure sensor, and the pressure deviation vector is calculated. The influence of the current temperature of the workbench on the pressure loss is read, and the maximum allowable displacement of the fine-tuning actuator is used as a constraint to solve the objective function to obtain the compensation displacement command vector. To avoid sudden changes in the command, the displacement command obtained by the solution is smoothed, and a feedforward compensation term is added in combination with the system dynamic response to obtain the final partitioned compensation displacement command. Active deformation execution module: The partition compensation displacement command is synchronously applied to each fine-tuning actuator through digital communication to stretch or shrink to the command position, so as to actively adapt to the current back surface topography of the wafer.

6. The control system for a fully automated probe platform according to claim 5, characterized in that, The contact event sensing and processing module is communicatively connected to the signal output interface of the test machine. The core control mode switching module is communicatively connected to the main servo driver in the z-axis motion component that drives the vertical movement of the worktable. The pressure distribution calculation and compensation instruction generation module is digitally connected to both the pressure sensor array and the fine-tuning actuator array.

7. A control method for a fully automated probe platform, implemented by a control system for a fully automated probe platform as described in any one of claims 1-6, characterized in that, Includes the following steps: S1. Contact Initiation Event Determination: During the movement of the fully automated probe stage toward the probe card, the electrical feedback signal from the test machine is continuously monitored. By identifying that the current change rate of the electrical feedback signal exceeds the preset threshold calibrated based on the background noise, the contact initiation event between the probe and the wafer test pad is determined in real time. S2, Control Mode Switching: After determining the contact initiation event, the control mode of the servo axis on the main servo drive that drives the vertical movement of the worktable is immediately switched from position control mode to direct force control mode with the goal of maintaining the preset total contact pressure. S3. Pressure distribution calculation and command generation: In direct force control mode, the readings of the independent pressure sensor array arranged under the workbench bearing surface are collected synchronously. Based on the pre-calibrated relationship matrix that reflects the influence of the displacement of each fine-tuning actuator on the pressure of each pressure sensor, the deviation between each pressure reading and the preset total contact pressure is calculated into a set of partition compensation displacement commands for driving the corresponding area fine-tuning actuator. S4. Active Deformation Execution: The partition compensation displacement command is synchronously applied to each fine-tuning actuator, so that the bearing surface of the worktable generates controlled elastic deformation through the deformation of the flexible support body below it, so as to actively adapt to the back surface morphology of the current wafer and promote the uniformity of the contact pressure of each probe on the probe card.

8. A fully automated probe platform, employing the control system for a fully automated probe platform according to any one of claims 1-6, and the control method for a fully automated probe platform according to claim 7, characterized in that, Including industrial control power distribution box (1); The electrical control system (2) is integrated inside the industrial control power distribution box (1), and the vision system (9) is installed on the upper surface of the industrial control power distribution box (1) and connected to the electrical control system (2). The human-machine interface (3), the flexible positioning support frame (5) and the picking and placing robot arm (6) are movably installed on the upper surface of the industrial control power distribution box (1) from left to right. A precision positioning four-dimensional motion module (4) is set in the middle of the flexible positioning support frame (5) and inside the industrial control power distribution box (1). One end of the picking and placing robot arm (6) is movably connected to the basket loading position (7). A pre-alignment component (8) is set between the picking and placing robot arm (6) and the basket loading position (7). The precision positioning four-dimensional motion module (4) includes a detection tray (401), a worktable (406) is provided in the middle of the upper surface of the detection tray (401), and the side of the worktable (406) is provided with an x-axis motion component (402), a y-axis motion component (403), a z-axis motion component (404) and a θ-rotation axis motion component (405) in sequence from bottom to top, and a probe card (407) is placed on the upper surface of the worktable (406).

9. A fully automated probe platform according to claim 8, characterized in that, The workbench (406) is provided with a partitioned fine-tuning structure, which includes a rigid bearing plate; A flexible support body is a metal structure connected to the bottom of the rigid bearing plate and equipped with flexible hinges or weak stiffeners. A fine-tuning actuator array is arranged between the flexible support and the z-axis motion assembly (404); A pressure sensor array is configured to measure the support reaction force of each partition of the partitioned fine-tuning structure.

10. A fully automated probe platform according to claim 9, characterized in that, The fine-tuning actuator array is a piezoelectric ceramic actuator array or a voice coil motor array; The pressure sensor array is a strain gauge sensor array, integrated into the flexible support body.

Images