Semiconductor test probe card based on vertical electric shielding structure and manufacturing method thereof

Abstract

The invention relates to the technical field of semiconductor testing, in particular to a semiconductor testing probe card based on a vertical electric shielding structure and a manufacturing method thereof, the probe card comprises a substrate and a probe array arranged on the substrate, and further comprises a control module electrically connected with the probe array, and the control module comprises a processing unit and a storage unit; a dynamic characteristic model is pre-stored in the storage unit, and the dynamic characteristic model is used for representing leakage current characteristics of the probe card in different working states. According to the invention, feedforward prediction and active compensation are carried out through a software algorithm, and the beneficial effects of eliminating dynamic leakage current errors in real time and improving the measurement precision are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of semiconductor testing, and in particular to a semiconductor test probe card based on a vertical electric shielding structure and a manufacturing method thereof. BACKGROUND

[0002] In the process of electrical parameter testing of a semiconductor wafer or chip, a probe card, as a key interface connecting a test device and a device under test, directly affects the accuracy of test results due to its own electrical characteristics. Leakage current generated by the probe card is an important factor causing errors in weak current measurement. In order to suppress leakage current, the prior art usually seeks solutions from two directions of hardware structure optimization or passive monitoring.

[0003] However, the leakage current characteristics of the probe card of the prior art are affected in real time by various dynamic working parameters when it is working, including but not limited to voltage changes on adjacent probes, working frequency of signals, and working temperature fluctuations of the probe card itself. Therefore, the prior art solution cannot effectively eliminate dynamic leakage current errors caused by real-time changes of dynamic working parameters in the actual testing process, which will cause the problem of insufficient test precision when high-precision and low-current measurement is performed. SUMMARY

[0004] The present application aims to provide a semiconductor test probe card based on a vertical electric shielding structure, comprising a substrate and a probe array arranged on the substrate, further comprising: a control module electrically connected to the probe array, the control module comprising a processing unit and a storage unit; the storage unit pre-stores a dynamic characteristic model, the dynamic characteristic model being used to represent leakage current characteristics of the probe card under different working states; the processing unit is configured to, in the process of performing electrical parameter testing on a device under test: acquire at least one dynamic working parameter affecting leakage current in real time; input the dynamic working parameter into the dynamic characteristic model to calculate a predicted leakage current value corresponding to the current testing condition; receive a measured current value of the device under test from a measurement unit of a test device; calculate a corrected current value by subtracting the predicted leakage current value from the measured current value; and output the corrected current value as a final test result to actively compensate for leakage current errors caused by changes in the dynamic working parameter.

[0005] By adopting the above technical solution, in the testing process, the processing unit acquires dynamic working parameters affecting leakage current in real time, calculates a predicted leakage current value under the current condition by using the model, and subtracts the predicted value from a measured current value measured by a test device, so as to obtain a corrected current value as a final test result. Through software algorithm for feedforward prediction and active compensation, the beneficial effects of real-time elimination of dynamic leakage current errors and improvement of measurement precision are achieved.

[0006] Optionally, the dynamic characteristic model is a neighborhood-aware crosstalk leakage prediction model, which establishes a functional mapping relationship between the leakage current of the measured probe and the electrical state of at least one neighboring probe of the measured probe; the dynamic operating parameters obtained by the processing unit include the real-time voltage or real-time operating frequency of the at least one neighboring probe.

[0007] By adopting the above technical solution, the dynamic leakage component generated by the adjacent channel crosstalk can be accurately predicted and quantified, thereby realizing targeted compensation for the main error source.

[0008] Optionally, the dynamic operating parameters obtained by the processing unit further include real-time temperature data collected by a temperature sensor arranged on the probe card; the neighborhood-aware crosstalk leakage prediction model further establishes a functional mapping relationship between the leakage current of the measured probe and the real-time temperature data.

[0009] By adopting the above technical solution, the compensation system can adapt to the change of the thermal environment in the test process, thereby maintaining high measurement accuracy in a wider operating temperature range.

[0010] Optionally, it further includes a calibration management module, which is configured to: under a preset triggering condition, automatically execute a standardized probe card characteristic characterization process; based on the measurement results of the characteristic characterization process, generate a feature fingerprint data file containing parameters required by the neighborhood-aware crosstalk leakage prediction model; bind the feature fingerprint data file with the unique identity identifier of the probe card, and store it in the storage unit for calling by the processing unit when performing the test.

[0011] By adopting the above technical solution, the whole life cycle automatic management of model parameters is realized, and it is ensured that the compensation algorithm always maintains high precision in the entire use cycle of the probe card from new to aging.

[0012] Optionally, the parameters contained in the feature fingerprint data file at least include: baseline leakage parameters representing the inherent leakage of each probe; a crosstalk coupling coefficient matrix representing the strength of the electric field coupling between the neighboring probes; and at least one temperature-dependent coefficient representing the change of the leakage current with temperature.

[0013] By adopting the above technical solution, a data basis is provided for constructing a robust model that can simultaneously consider the probe's own properties, neighborhood crosstalk effect and temperature influence.

[0014] Optionally, the characteristic characterization procedure performed by the calibration management module comprises the following steps: performing a baseline leakage measurement, i.e. measuring and recording the inherent leakage current of each probe when all adjacent probes are in a preset non-energized state to determine the baseline leakage parameter; performing a crosstalk matrix scan, i.e. systematically applying different voltage or frequency excitation signals to the designated adjacent probes and synchronously measuring the leakage current change on the measured probe to calculate and generate the crosstalk coupling coefficient matrix; performing a temperature characteristic scan, i.e. repeating the key leakage current measurement in a controlled temperature variation range to fit the temperature-dependent coefficient.

[0015] By adopting the technical solutions described above, a complete and rigorous calibration methodology is provided, which ensures that the generated characteristic fingerprint data file has high consistency and reliability.

[0016] Optionally, the preset trigger condition comprises at least one of the following: first installation and initialization of the probe card, periodic maintenance after reaching a preset number of test cycles or working time, or after completing physical repair or component replacement.

[0017] By adopting the technical solutions described above, it is ensured that the characteristic model of the probe card is updated in time at key nodes where the state of the probe card may change significantly, thereby ensuring the effectiveness of compensation.

[0018] Optionally, the functional logic of the processing unit is integrated into the underlying measurement control software of the test equipment, so that the calculation of the predicted leakage current value and the generation of the corrected current value can be completed in real time without significantly increasing the test time overhead.

[0019] By adopting the technical solutions described above, the calculation of the predicted leakage current value and the generation of the corrected current value can be completed in real time and seamlessly with the measurement hardware, thereby achieving the improvement of precision without sacrificing test throughput.

[0020] Optionally, the control module is further configured to automatically find and load the dynamic characteristic model or its parameter file corresponding to the installed probe card from the central database to the storage unit according to the unique identity identifier of the installed probe card before the test starts.

[0021] By adopting the technical solutions described above, the efficiency and reliability of production and operation are improved.

[0022] The second object of the present application is to provide a manufacturing method of the semiconductor test probe card with vertical electric shielding structure, comprising: providing a substrate with vertical electric shielding structure, and assembling a probe array thereon to form a probe card hardware body; integrating a control module containing a processing unit and a storage unit on the probe card hardware body; performing a one-time initialization characterization process, which includes: measuring the baseline leakage parameters of each probe; calculating the crosstalk coupling coefficient matrix by applying excitation to adjacent probes and measuring the response; and determining the temperature dependence coefficient by measuring at different temperatures; generating an initial feature fingerprint data file according to the results of the characterization process; and storing the initial feature fingerprint data file in the storage unit, thereby completing the manufacturing and factory calibration of the probe card.

[0023] By adopting the above technical solution, a complete manufacturing and factory calibration process is established, ensuring that each product shipped from the factory is pre-loaded with a personalized and accurate dynamic compensation model, which can provide high-precision test performance from the first use. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 The system block diagram of the semiconductor test probe card with vertical electric shielding structure according to the present application.

[0025] Figure 2 The manufacturing method flowchart of the semiconductor test probe card with vertical electric shielding structure according to the present application. DETAILED DESCRIPTION

[0026] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described in detail below in conjunction with specific embodiments. Obviously, the described embodiments are some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0027] In order to solve the problem that the semiconductor test probe card in the prior art cannot adaptively compensate for the leakage current error caused by dynamic working parameter changes such as adjacent probe electrical activity and working temperature fluctuation, thereby resulting in insufficient test precision in high-precision, low-current measurement scenarios, the present application establishes a mathematical model for each probe card that can accurately characterize its dynamic leakage characteristics, and uses this model to calculate and subtract the dynamically generated leakage error component in real time during the test process, thereby obtaining a corrected and more true electrical parameter value of the device under test.

[0028] As Figure 1As shown, the present application provides a semiconductor test probe card based on vertical electric shielding structure, which comprises a substrate and a probe array arranged on the substrate. The substrate can be a multilayer ceramic substrate or a high-performance organic substrate, which is internally designed with a precise vertical electric shielding structure, such as a ground shielding via array or a coaxial shielding layer arranged around the signal via, to provide good electromagnetic isolation for the probe signal via from a physical level, so as to suppress static and partial dynamic crosstalk. The probe array can be composed of thousands or even tens of thousands of probes, which can be cantilever beam probes, vertical probes or other types of micro-electro-mechanical system probes, for establishing reliable electrical contact with the pads on the semiconductor wafer or chip under test.

[0029] It can be understood that the probe card of the present application further comprises a control module electrically connected to the probe array, wherein the control module comprises a processing unit and a storage unit. The storage unit pre-stores a dynamic characteristic model, which is used to represent the leakage current characteristics of the probe card under different working conditions. The processing unit is configured to, during the electrical parameter test of the device under test: acquire at least one dynamic working parameter affecting the leakage current in real time; input the dynamic working parameter into the dynamic characteristic model to calculate the predicted leakage current value corresponding to the current test condition; receive the measured current value of the device under test from the measurement unit of the test equipment; calculate the corrected current value by subtracting the predicted leakage current value from the measured current value; and output the corrected current value as the final test result to actively compensate for the leakage current error caused by the change of the dynamic working parameter.

[0030] Specifically, the control module can be physically realized as a special integrated circuit integrated on the main printed circuit board of the probe card, or composed of a high-performance field programmable gate array or a microcontroller unit with on-chip peripherals. The control module is connected to the probe array and the test equipment interface through the wiring on the probe card, so as to monitor and control the signals in the test process.

[0031] It can be understood that the processing unit can be a central processing unit kernel in a microcontroller, or a special digital signal processing logic realized in a field programmable gate array, for high-speed mathematical operation according to a preset algorithm process.

[0032] The storage unit is used to store the model and parameters relied on by the compensation algorithm in a persistent manner, and the storage unit preferably adopts a non-volatile storage medium, such as a flash memory or an electrically erasable programmable read-only memory; the reason for adopting the non-volatile memory is that the dynamic characteristic model and the related parameters stored therein are specific to each probe card, and need to be completely saved when the probe card is powered off, transported or transferred between different test devices; the storage unit not only stores the algorithm itself, but more importantly, stores the data generated by the characteristic representation process which will be described below.

[0033] It can be understood that the dynamic characteristic model is not a static offset or a simple lookup table, but a multivariate function which mathematically describes how the leakage current of the probe card changes with one or more dynamically changing dynamic operating parameters; the input of the dynamic characteristic model is the real-time value of the dynamic operating parameters, and the output is the leakage current component generated by the probe card itself and superimposed on the real signal of the device under test under the joint action of these parameters.

[0034] The dynamic operating parameters refer to physical quantities which change in real time during the test and have a significant impact on the leakage current, and these dynamic operating parameters at least include the voltage state of the adjacent probe, the operating frequency of the signal on the adjacent probe and the real-time temperature of the environment in which the probe card is located; it can be understood that the range of the dynamic operating parameters is not limited to this, and in some more demanding applications, it can also include the voltage fluctuation of the test power supply, the signal slew rate and even the environmental humidity.

[0035] The processing unit acquires these dynamic operating parameters in real time in various ways, for example, for the voltage state of the adjacent probe, the processing unit can monitor the control instructions sent by the test equipment to each probe channel through the internal bus, so as to know that each adjacent probe is currently set to high level, low level, high impedance state or is outputting a dynamic signal; for the operating frequency, it can also be obtained by analyzing the clock information in the test vector; for the temperature, it is obtained by reading the temperature sensor data integrated on the probe card.

[0036] After acquiring the current values of all related dynamic operating parameters, the processing unit substitutes these values as inputs into the dynamic characteristic model stored in the storage unit for calculation, and this calculation process is essentially a function evaluation process, and the result is a specific current value, i.e. a predicted leakage current value, which represents the size of the leakage current that the probe card itself is expected to generate under the current specific time, specific electrical and temperature environment.

[0037] The measurement unit of the test equipment can be a parameter measurement unit or a source measurement unit, which performs actual current measurement on the corresponding pin of the device under test. The measurement result is the measured current value, which can be understood as a mixed value containing the real current of the device under test and the leakage current generated by the probe card.

[0038] The processing unit receives the measured current value from the measurement unit of the test equipment, and then calculates the corrected current value by subtracting the predicted leakage current value from the measured current value. The known error component is accurately stripped from the total measurement value containing the signal and the error, and the real signal value is purified.

[0039] Finally, the processing unit outputs the corrected current value as the final test result, which can be directly transmitted to the test program for comparison with the preset pass or fail standard, or recorded in the test log. At the moment of measurement, the error is calculated and eliminated in advance, so that each data point output to the test system is corrected with high precision, greatly improving the accuracy and reliability of weak current measurement, such as nanoampere or picoampere level, in complex dynamic environment.

[0040] It can be understood that the compensation logic can be a bottom-layer software driver or firmware function. When the test program issues a measurement instruction, the hardware controller of the test equipment obtains the measured current value, which is automatically intercepted by the bottom-layer system before being returned to the upper-layer test program. The bottom-layer system simultaneously reads all relevant dynamic working parameters from the state register of the test machine and calls the compensation function for calculation and correction. The whole process is completed within microseconds and is completely transparent to the upper-layer test program.

[0041] It can be understood that the control module is further configured to automatically find and load the dynamic characteristic model or the parameter file thereof corresponding to the installed probe card from the central database to the storage unit according to the unique identity identifier of the probe card before the test starts; in this scheme, each probe card is assigned a globally unique identity identifier at the time of manufacture, such as a serial number burned in a one-time programmable memory or a radio frequency identification tag attached to the circuit board; the central database is maintained on the server of the factory, and stores the characteristic fingerprint data files of all probe cards and corresponds to the unique identity identifiers one by one; when a probe card is installed on any test equipment, the host software of the test equipment first reads the identity identifier of the card; then, the software initiates a query request to the central database through the factory local area network; the database finds the matching characteristic fingerprint data file according to the received identity identifier and sends it back to the test equipment; the test equipment software immediately downloads this file to the storage unit of the control module of the probe card, or loads it into the memory area reserved by the test equipment itself for compensation algorithm; completely eliminates the risk of test result deviation caused by manual selection of wrong calibration files, and improves the operation and maintenance efficiency and data reliability of the production line.

[0042] It can be understood that the dynamic characteristic model can be a neighborhood-aware crosstalk leakage prediction model, which establishes a functional mapping relationship between the leakage current of the measured probe and the electrical state of at least one adjacent probe of the measured probe; the dynamic working parameters obtained by the processing unit include the real-time voltage or real-time working frequency of at least one adjacent probe.

[0043] It can be understood that the above-mentioned functional mapping relationship can have various mathematical implementation forms, one of which is to use a linear superposition model; assuming that for a measured probe i, its total leakage current I leaktotal (i) is composed of the inherent baseline leakage I base (i) and the sum of all crosstalk components from adjacent probes, for each adjacent probe j, its crosstalk contribution to probe i can be approximated as a function of its electrical state, for example, proportional to its applied voltage V(j); therefore, the entire model can be expressed as: I leaktotal (i) = I base(i) +∑[C(i,j) * f(V(j), Freq(j))], where the summation sign ∑ traverses all neighboring probes j that have a significant influence on the probe i under test, f(V(j), Freq(j)) is a function that characterizes the electrical state of the neighboring probe j, which in many DC or low frequency test scenarios can be simplified to its real-time voltage V(j); C(i,j) is a crucial parameter, called the crosstalk coupling coefficient, quantifying how much a unit voltage change of the neighboring probe j will induce a leakage current change in the probe i under test, which depends on the distance between the two probes, the dielectric material properties between them, and the surrounding shielding structure design.

[0044] Because the coupling between probes has not only a resistive component but also a capacitive component, when testing high frequency signals, the electrical state also needs to consider the real-time operating frequency Freq(j); as the frequency increases, the capacitive reactance decreases, and the displacement current generated through the parasitic capacitance becomes an important component of the leakage current; therefore, in more complex models, the function f can contain frequency-dependent terms, or the crosstalk coupling coefficient C(i,j) itself is a function of frequency.

[0045] In order to further improve the accuracy of the model to cope with the inevitable temperature changes during the test, the present application also introduces temperature compensation; the dynamic operating parameters obtained by the processing unit also include real-time temperature data collected by the temperature sensor arranged on the probe card; the neighborhood-aware crosstalk leakage prediction model further constructs a functional mapping relationship between the leakage current of the probe under test and the real-time temperature data.

[0046] The leakage characteristics of semiconductor materials and insulating materials are usually very sensitive to temperature, and usually increase exponentially with temperature, so the dynamic model can include the temperature variable; the temperature sensor can be a negative temperature coefficient thermistor or a digital output integrated circuit temperature sensor, which can be placed on the probe card in a position that best represents the working temperature of the probe array region, such as the center of the probe array or near the hot spot region; the processing unit will periodically read the value of the sensor to obtain real-time temperature data.

[0047] Integrating temperature data into the neighborhood-aware crosstalk leakage prediction model can be achieved by making the key parameters of the model a function of temperature; for example, the baseline leakage I base (i) and the crosstalk coupling coefficient C(i,j) are no longer constants, but functions of temperature T, i.e. I base (i,T) and C(i,j,T); these two functional relationships can be determined by measuring at different temperatures and performing curve fitting; for example, the change of baseline leakage with temperature can often be described by the Arrhenius equation, i.e. I base(i,T) = A * exp(-Ea / (k * T)), where A is a proportionality factor, Ea is the activation energy, and k is the Boltzmann constant. In practical applications, this relationship can also be approximated by a polynomial function.

[0048] Therefore, the final form of the neighborhood-aware crosstalk leakage current prediction model considering both neighborhood crosstalk and temperature effects can be conceptually expressed as: I leaktotal (i) = I base (i,T) + Σ [C(i,j,T) * f(V(j), Freq(j))], during the test process, the processing unit obtains the voltage V(j) of the adjacent probe, the frequency Freq(j) and the real-time temperature T of the probe card in real time, and substitutes all these dynamic working parameters into the comprehensive model, so as to calculate the predicted leakage current value which can reflect all the main dynamic influencing factors at present.

[0049] With the use of the probe card, the probes will wear out and the insulating material will age, and these physical changes will cause the leakage characteristics to slowly drift, thereby making the initially established model parameters gradually inaccurate. In order to solve this problem, the application designs a closed-loop automatic calibration management mechanism.

[0050] Specifically, it also includes a calibration management module configured to: automatically execute a standardized probe card characteristic characterization process under a preset trigger condition; generate a feature fingerprint data file containing the parameters required by the neighborhood-aware crosstalk leakage prediction model based on the measurement results of the characteristic characterization process; bind the feature fingerprint data file with the unique identifier of the probe card, and store it in the storage unit for the processing unit to call when performing the test.

[0051] The calibration management module monitors the usage state of the probe card and automatically starts or requests the test system to start a re-calibration process when certain conditions are met.

[0052] In order to enable this calibration process to cover all the information required by the model, the application specifies the content of the data file generated after calibration, for example, the parameters contained in the feature fingerprint data file at least include: baseline leakage parameters representing the inherent leakage of each probe; crosstalk coupling coefficient matrix representing the strength of the electric field coupling between adjacent probes; and at least one temperature-dependent coefficient representing the change of leakage current with temperature.

[0053] Specifically, the baseline leakage parameter is a set of data recording the inherent leakage current value of each probe itself at a standard temperature, for example twenty-five degrees Celsius, when all adjacent probes are in a non-energized state, such as grounded or high resistance state.

[0054] The crosstalk coupling coefficient matrix is a two-dimensional array or matrix, if the probe card has N probes, it is theoretically an N x N matrix, where the matrix element C(i,j) represents the strength of the influence of the excitation of probe j on the leakage current of probe i; in practice, due to the locality of the crosstalk effect, there is significant coupling only between physically very close probes, so this matrix is usually a sparse matrix, only the non-zero elements need to be stored, thereby greatly saving storage space.

[0055] The temperature-dependent coefficients are a set of parameters describing how the baseline leakage and crosstalk coupling coefficients change with temperature; for example, for baseline leakage, it can be the activation energy Ea in the Arrhenius equation; for crosstalk coefficients, it can be a linear temperature coefficient a, these coefficients enable the model to make accurate predictions at any operating temperature.

[0056] Packaging the baseline leakage parameters, crosstalk coupling coefficient matrix, and temperature-dependent coefficients together forms a complete feature fingerprint data file, and is firmly bound with the unique identifier of the probe card, stored in the local storage unit of the probe card or the central database of the factory.

[0057] It can be understood that the preset triggering condition includes at least one of the following: the first installation initialization of the probe card, the periodic maintenance after reaching the preset test cycle number or working time length, or after completing physical repair or component replacement.

[0058] Specifically, when a brand new or transferred to a new test equipment probe card is installed, the test system recognizes that it is a new combination, and automatically triggers a complete characterization process to establish the most accurate initial model.

[0059] The calibration management module maintains one or more counters inside, such as recording the total number of contacts of the probe or the cumulative power-on working time; when any counter reaches the preset threshold, the calibration management module requests the test system to re-calibrate; the test engineer can perform the calibration process at a convenient time.

[0060] When the probe card has undergone physical repair, such as replacing a part of the damaged probe, or replacing a certain element on the circuit board, its electrical characteristics are likely to have changed significantly; in this case, a complete recalibration needs to be performed to ensure that the new physical state can be accurately reflected in the feature fingerprint data file.

[0061] It can be understood that the characteristic characterization procedure performed by the calibration management module includes the following steps: performing a baseline leakage measurement, i.e. measuring and recording the inherent leakage current of each probe to determine the baseline leakage parameter when all adjacent probes are in a preset non-excitation state; performing a crosstalk matrix scan, i.e. systematically applying different voltage or frequency excitation signals to the designated adjacent probe and synchronously measuring the leakage current change on the measured probe to calculate and generate a crosstalk coupling coefficient matrix; performing a temperature characteristic scan, i.e. repeating the key leakage current measurement in a controlled temperature variation range to fit the temperature-dependent coefficient.

[0062] Specifically, through the channel controller of the test equipment, all other probes in the probe array except one measured probe are set to a preset non-excitation state to create the cleanest electromagnetic environment; then, the parameter measurement unit of the test equipment applies a standard test voltage to the measured probe and measures the tiny current flowing through the probe; this current value is the baseline leakage of the probe at the current temperature; then, the next probe is automatically switched as the measured probe and the above process is repeated until the baseline leakage of each probe in the probe array is measured and recorded; finally, these data are processed to form the baseline leakage parameter part in the feature fingerprint data file.

[0063] It can be understood that the program will again traverse each probe to select it as the measured probe i, and for each measured probe i, the program will further traverse all physically adjacent probes j thereof; for each adjacent probe j, the following operations will be performed: first, keeping all other probes in a non-excitation state, applying a known excitation signal, such as a 1V DC voltage, to the adjacent probe j; at the same time, the parameter measurement unit of the test equipment continuously monitors the leakage current on the measured probe i; due to the excitation of the adjacent probe j, the leakage current on the measured probe i will change by a small amount ΔI leak (i); the crosstalk coupling coefficient C(i,j) can be calculated by this change amount, for example, C(i,j) = ΔI leak (i) / V stimulus (j). If the frequency effect needs to be considered, then the excitation signal applied can be an alternating current signal of a specific frequency, and the frequency-dependent coupling coefficient is calculated by measuring the alternating current response on the measured probe; finally, all calculated C(i,j) values are collected together to form the crosstalk coupling coefficient matrix.

[0064] It can be understood that the probe card can be placed in an environment that can precisely control the temperature, such as a temperature-controlled chuck or a small temperature box, at several different key temperature points, such as 25 degrees Celsius, 50 degrees Celsius, 75 degrees Celsius and 100 degrees Celsius, and a set of measurement values of each key parameter at different temperatures will be obtained for each key parameter, including the baseline leakage Ibase(i) of each probe and some typical crosstalk coupling coefficients C(i,j), etc., and finally, curve fitting is performed on these data points, such as using an exponential function to fit the baseline leakage data to extract the activation energy and other parameters in the Arrhenius equation; the crosstalk coefficient data is fitted using a linear or quadratic polynomial to obtain its temperature coefficient; the coefficients obtained by fitting are the temperature-dependent coefficients finally stored in the characteristic fingerprint data file.

[0065] As shown in Figure 2 The application provides a manufacturing method of the above-mentioned semiconductor test probe card, including the following steps: providing a substrate with a vertical shielding structure, and assembling a probe array on the substrate to form a probe card hardware main body; integrating a control module including a processing unit and a storage unit on the probe card hardware main body; performing a one-time initialization characteristic characterization process, which includes: measuring the baseline leakage parameters of each probe; calculating the crosstalk coupling coefficient matrix by applying excitation to adjacent probes and measuring the response; and determining the temperature-dependent coefficients by measuring at different temperatures; generating an initial characteristic fingerprint data file according to the results of the characteristic characterization process; and storing the initial characteristic fingerprint data file in the storage unit, thereby completing the manufacturing and factory calibration of the probe card.

[0066] Specifically, a multi-layer substrate with a complex internal vertical shielding structure is first manufactured according to design drawings, and then a probe array composed of thousands of tiny probes is installed on the substrate through a precision mounting process to form the core hardware of the probe card.

[0067] On the main printed circuit board of the probe card, a control module chip including a processing unit and a storage unit, a temperature sensor, and other necessary interface and power management circuits are soldered through surface mounting technology to complete the manufacturing of the probe card hardware main body.

[0068] The just assembled probe card is installed on a factory-specific calibration test platform, which is configured to accurately perform the complete characteristic characterization process described above. The test platform will automatically perform a comprehensive inspection of this new probe card, including systematically measuring the baseline leakage of all probes at multiple temperature points and scanning a complete crosstalk coupling coefficient matrix.

[0069] The calibration test platform processes and calculates the mass measurement data collected in the previous step, extracts all baseline leakage parameters, crosstalk coupling coefficients and temperature dependence coefficients, and organizes these parameters in a predetermined format to generate an initial characteristic fingerprint data file specific to this probe card.

[0070] Finally, the calibration test platform permanently writes the initial characteristic fingerprint data file just generated, together with the unique identity identifier of the probe card, into the non-volatile storage unit of the probe card control module through a program interface.

[0071] At this point, the low-leakage semiconductor test probe card with dynamic compensation capability has completed manufacturing and factory calibration, carries its own accurate electrical characteristic digital archive, and can provide high-precision test results with active compensation from the first use.

[0072] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Various changes and modifications can be made by those skilled in the art based on the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0073] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is taken as an example for illustration, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0074] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules or units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed each other can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0075] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e. may be located in one place, or may be distributed on multiple network units. Part or all of the units may be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0076] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present alone, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0077] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part of the prior art that contributes to the technical solutions or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes a U disk, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk, and various program code storage media.

[0078] The above are the preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in the specification (including the abstract and the drawings) can be replaced by other equivalent or similar features unless specifically stated. That is, each feature is only an example of a series of equivalent or similar features unless specifically stated.

Claims

1. A semiconductor test probe card based on a vertical electrically shielded structure, comprising a substrate and a probe array disposed on the substrate, characterized in that, Also includes: A control module electrically connected to the probe array, the control module comprising a processing unit and a storage unit; The storage unit pre-stores a dynamic feature model, which is used to characterize the leakage current characteristics of the probe card under different working states. The processing unit is configured to, during the process of performing electrical parameter testing on the device under test: Real-time acquisition of at least one dynamic operating parameter that affects leakage current; The dynamic operating parameters are input into the dynamic feature model to calculate the predicted leakage current value corresponding to the current test conditions; Receive the measured current value for the device under test from the measurement unit of the test equipment; The corrected current value is calculated by subtracting the predicted leakage current value from the measured current value. as well as The corrected current value is output as the final test result to actively compensate for leakage current errors caused by changes in the dynamic operating parameters.

2. The low-leakage semiconductor test probe card according to claim 1, characterized in that, The dynamic feature model is a neighborhood-aware crosstalk leakage current prediction model, which constructs a functional mapping relationship between the leakage current of the probe under test and the electrical state of at least one neighboring probe of the probe under test; the dynamic operating parameters obtained by the processing unit include the real-time voltage or real-time operating frequency of the at least one neighboring probe.

3. The low-leakage semiconductor test probe card according to claim 2, characterized in that, The dynamic operating parameters acquired by the processing unit also include real-time temperature data collected by a temperature sensor set on the probe card; the neighborhood-aware crosstalk leakage current prediction model further constructs a functional mapping relationship between the leakage current of the probe under test and the real-time temperature data.

4. The low-leakage semiconductor test probe card according to claim 2, characterized in that, It also includes a calibration management module, which is configured to: Under preset trigger conditions, the standardized probe card characteristic characterization process is automatically executed; Based on the measurement results of the aforementioned characteristic characterization process, a feature fingerprint data file containing the parameters required by the neighborhood-aware crosstalk leakage current prediction model is generated. The feature fingerprint data file is bound to the unique identifier of the probe card and stored in the storage unit for the processing unit to access during testing.

5. The low-leakage semiconductor test probe card according to claim 4, characterized in that, The parameters included in the feature fingerprint data file include at least: baseline leakage current parameters characterizing the inherent leakage current of each probe; crosstalk coupling coefficient matrix characterizing the electric field coupling strength between adjacent probes; and at least one temperature dependence coefficient characterizing the leakage current as a function of temperature.

6. The low-leakage semiconductor test probe card according to claim 4, characterized in that, The characteristic characterization process executed by the calibration management module includes the following steps: Perform baseline leakage current measurement, that is, measure and record the inherent leakage current of each probe when all neighboring probes are in a preset non-excitation state to determine the baseline leakage current parameters; Perform crosstalk matrix scanning, which involves systematically applying different voltage or frequency excitation signals to designated neighboring probes and simultaneously measuring the leakage current change on the probe under test in order to calculate and generate the crosstalk coupling coefficient matrix. Perform a temperature characteristic scan, which involves repeating critical leakage current measurements within a controlled temperature range to fit the temperature dependence coefficient.

7. The low-leakage semiconductor test probe card according to claim 4, characterized in that, The preset triggering conditions include at least one of the following: the initial installation and initialization of the probe card, periodic maintenance after reaching a preset number of test cycles or working hours, or after physical repair or component replacement.

8. The low-leakage semiconductor test probe card according to claim 1, characterized in that, The functional logic of the processing unit is integrated into the underlying measurement and control software of the test equipment, thereby ensuring that the calculation of the predicted leakage current value and the generation of the corrected current value can be completed in real time without significantly increasing the test time overhead.

9. The low-leakage semiconductor test probe card according to claim 1, characterized in that, The control module is further configured to: before the test begins, automatically search for and load the corresponding dynamic feature model or its parameter file from the central database to the storage unit based on the unique identifier of the installed probe card.

10. A method for manufacturing a semiconductor test probe card as described in any one of claims 1-9, characterized in that, include: A substrate with a vertical electrical shielding structure is provided, and a probe array is assembled on it to form the hardware body of the probe card; A control module containing a processing unit and a storage unit is integrated on the probe card hardware body; Perform an initial characterization process, which includes: measuring the baseline leakage parameters of each probe; calculating the crosstalk coupling coefficient matrix by applying excitation to neighboring probes and measuring the response; and determining the temperature dependence coefficient by performing measurements at different temperatures. Based on the results of the aforementioned feature characterization process, an initial feature fingerprint data file is generated; and The initial feature fingerprint data file is stored in the storage unit, thereby completing the manufacturing and factory calibration of the probe card.

Images