Method and system for electrical performance testing of a semiconductor probe apparatus

Abstract

The application provides a kind of electrical performance test method and system of semiconductor probe equipment, including the first excitation signal containing first direct current bias, first frequency sine wave is applied to equipment multi-probe, the first response signal is collected, the fundamental wave contact resistance of each probe is calculated, the first direct current bias step is adjusted, the steady-state recovery time of fundamental wave contact resistance is measured to determine the relaxation time of each probe;According to the relaxation time of each probe, the second frequency representing the characteristic is matched, the second excitation signal containing the second direct current bias and the corresponding second frequency sine wave is applied, the second response signal is collected, and the harmonic component amplitude of the signal at the preset integer multiple frequency of the second frequency is calculated;Single probe is taken as target probe, reference probe is screened from spatial neighborhood according to fundamental wave contact resistance similarity, and the harmonic component amplitude of the two is combined to construct the spatial correlation characteristic vector of target probe, and the electrical performance of equipment is judged based on the fundamental wave contact resistance of each probe and the corresponding spatial correlation characteristic vector.

Description

Technical Field

[0001] This application belongs to the field of performance testing, and in particular relates to a method and system for testing the electrical performance of a semiconductor probe device. Background Technology

[0002] Electrical performance testing methods typically focus on measuring probe parameters by applying DC current and measuring the voltage drop to calculate the contact resistance between each probe and the test pad. However, the electrical characteristics of the probe-pad contact interface are not ideal linear ohmic contacts, especially after wear, contamination, or deformation, exhibiting characteristics that are difficult to detect with DC testing. AC testing methods use one or a few fixed test frequencies for all probes, ignoring the response differences between probes due to variations in manufacturing tolerances, wear levels, and contamination conditions. The characteristics of some probes may only become apparent within specific frequency ranges, and fixed test frequencies may fail to excite and detect potential defects. Furthermore, probe failure or performance degradation may exhibit localized clustering characteristics; existing methods lack the ability to analyze such spatial clustering effects, making it difficult to distinguish between random defects and systematic regional problems, resulting in an incomplete assessment of the overall health of the probe equipment. Summary of the Invention

[0003] In response to the problems mentioned in the background art, in a first aspect, the present invention proposes an electrical performance testing method for a semiconductor probe device, comprising:

[0004] A first excitation signal comprising a first DC bias and a first frequency sine wave is applied to multiple probes of the semiconductor probe device, a corresponding first response signal is acquired, and the fundamental contact resistance of each probe is calculated based on the first response signal; a preset step change is applied to the first DC bias, the time required for the fundamental contact resistance to recover to a steady state is measured, and the relaxation time of each probe is determined.

[0005] Based on the relaxation time of each probe, a second frequency for representing the characteristics is determined for the probe; for each probe, a second excitation signal including a second DC bias and a sine wave of the second frequency determined for the probe is applied, a corresponding second response signal is acquired, and the amplitude of the harmonic components of the second response signal at at least a preset integer multiple of the second frequency is calculated.

[0006] For each probe as a target probe, at least one reference probe is selected in the probe's spatial neighborhood based on the similarity of the fundamental contact resistance between other probes in the neighborhood and the target probe. The spatial correlation feature vector is constructed for the target probe using the harmonic component amplitudes of the target probe and the reference probe.

[0007] The electrical performance of the semiconductor probe device is determined based on the fundamental contact resistance of each probe and the spatial correlation feature vector corresponding to the probe.

[0008] In a second aspect, the present invention proposes an electrical performance testing system for a semiconductor probe device, comprising the following modules:

[0009] The determination module is used to apply a first excitation signal containing a first DC bias and a first frequency sine wave to multiple probes of the semiconductor probe device, acquire the corresponding first response signal, and calculate the fundamental contact resistance of each probe based on the first response signal; apply a preset step change to the first DC bias, measure the time required for the fundamental contact resistance to recover to a steady state, and determine the relaxation time of each probe.

[0010] The calculation module is used to determine a second frequency for representing the characteristics of each probe based on the relaxation time of each probe; for each probe, apply a second excitation signal including a second DC bias and a sine wave of the second frequency determined for the probe, acquire the corresponding second response signal, and calculate the amplitude of the harmonic components of the second response signal at at least a preset integer multiple of the second frequency.

[0011] The construction module is used to select at least one reference probe in the probe's spatial neighborhood based on the similarity of the fundamental contact resistance of other probes in the neighborhood to the target probe, and to construct a spatial correlation feature vector for the target probe using the harmonic component amplitudes of the target probe and the reference probe.

[0012] The determination module is used to determine the electrical performance of the semiconductor probe device based on the fundamental contact resistance of each probe and the spatial correlation feature vector corresponding to the probe.

[0013] This invention analyzes the response recovery process of probes after a step change in excitation, determining specific test conditions for each probe to elicit the desired characteristics, thus improving the sensitivity for detecting early performance degradation or potential failures of probes. Utilizing spatial correlation analysis of probe performance, by constructing performance correlation characteristics between the target probe and probes in its neighborhood, it distinguishes between random defects and regional problems, providing insight into the overall health status of the probe equipment. Judgments based on the probe's fundamental contact resistance, harmonic response, and spatial correlation characteristics result in more comprehensive and reliable evaluation results. Attached Figure Description

[0014] Figure 1 A flowchart of the first embodiment;

[0015] Figure 2 This is a schematic diagram of the first excitation signal;

[0016] Figure 3 A schematic diagram showing the change of fundamental contact resistance over time and the determination of relaxation time;

[0017] Figure 4 A schematic diagram of the harmonic component frequencies of the response signal. Detailed Implementation

[0018] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0019] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0020] In the first embodiment, the present invention proposes a method for testing the electrical performance of a semiconductor probe device, such as... Figure 1 ,include:

[0021] S1, apply a first excitation signal including a first DC bias and a first frequency sine wave to multiple probes of the semiconductor probe device, acquire the corresponding first response signal, and calculate the fundamental contact resistance of each probe based on the first response signal; apply a preset step change to the first DC bias, measure the time required for the fundamental contact resistance to recover to a steady state, and determine the relaxation time of each probe.

[0022] A DC bias voltage of, for example, 100 millivolts is applied to the probe using a source measurement unit, and a sine wave with an amplitude of 10 millivolts and a frequency of 1 kilohertz is superimposed on this DC bias as the first excitation signal. Figure 2The current flowing through the probe and the voltage across the probe are synchronously acquired using a data acquisition card as the first response signal. Fast Fourier Transform is then performed on the acquired voltage and current signals to extract the voltage amplitude at a frequency of 1 kHz. and current amplitude Then the fundamental contact resistance of the probe Calculated as and The ratio of .

[0023] While maintaining a constant 1 kHz sinusoidal excitation, the first DC bias is instantaneously stepped from 100 mV to 200 mV using a source measurement unit. Starting from the moment of the step, the response signal is continuously acquired, and the fundamental contact resistance is calculated within each time window, for example, every 1 ms, thus obtaining a series of resistance values ​​that change over time. When the resistance value changes calculated over multiple consecutive windows are less than a preset threshold, for example, less than 1% of the stable value, the resistance is considered to have recovered to a steady state. The time from the step occurrence to the first entry into this steady-state region is recorded as the relaxation time.

[0024] In an optional embodiment, calculating the fundamental contact resistance of each probe based on the first response signal includes:

[0025] The first frequency sine wave in the first excitation signal is used as the input current signal I(t), and the AC voltage signal V(t) in the first response signal is collected.

[0026] The amplitudes of the input current signal and the acquired AC voltage signal at the first frequency were extracted using Fourier transform. and ;

[0027] According to Ohm's law, the magnitude of the fundamental contact resistance is calculated. .

[0028] Specifically, a composite current signal containing a DC bias and an AC component is injected into the target probe, where the AC component is a first frequency, for example... The sine wave. The mathematical expression for the input current signal is: Assuming AC amplitude The voltage is 1 mA. Simultaneously, a voltage acquisition module synchronously measures the voltage signal at the contact point between the probe and the test pad, i.e., the first response signal V(t).

[0029] Fourier transforms were performed on the recorded time-series data of the input current signal I(t) and the output voltage signal V(t) to convert them to the frequency domain for analysis. From the obtained spectrum, the signal at the first frequency was extracted. The signal amplitude at that location. For example, extracting the input current amplitude from the current spectrum. The corresponding output voltage amplitude is extracted from the voltage spectrum in milliamperes. Millivolts. Based on Ohm's law, dividing the voltage amplitude by the current amplitude yields the fundamental contact resistance modulus of the probe at that frequency. Based on the example data, the calculation result is: .

[0030] To represent the response speed at which the probe contact interface recovers to a stable state after being subjected to an electrical disturbance, in an optional embodiment, the measurement of the time required for the fundamental contact resistance to recover to a steady state, determining the relaxation time of each probe, includes:

[0031] At any moment Apply a DC bias to the first DC bias by to The step voltage;

[0032] From time Initially, the fundamental contact resistance sequence was continuously measured and recorded at a fixed sampling period T. ;

[0033] Calculate the absolute value of the difference between adjacent measurements in the sequence. When the absolute value is less than a preset steady-state criterion threshold for a preset number of consecutive times At this time, record the current time as ;

[0034] The relaxation time Determined as .

[0035] At the initial moment The control system applies a sudden change to the DC bias voltage applied to the probe, for example, from a lower voltage. A step jump to a higher voltage A voltage step of volts. This voltage step instantaneously alters the electrochemical state of the contact interface, causing a transient change in contact resistance that gradually tends towards a new stable value.

[0036] from Starting from a certain point, the fundamental contact resistance is measured repeatedly at fixed time intervals, such as a sampling period T = 10 ms, thereby obtaining a time series data of the resistance value changing over time. For example, the sequence might exhibit resistances of 5.51 ohms, 5.35 ohms, 5.24 ohms, until it stabilizes. To determine when steady state is reached, the absolute value of the difference between the resistance values ​​at every two adjacent sampling points is continuously calculated. When the difference is less than a preset minimum value, such as a steady-state criterion threshold, for three consecutive times. When the resistance is 1 ohm, it is considered that the resistance value has stabilized. At this point, the first moment that the condition is met is recorded as... Assuming The relaxation time of the probe is then 50 μs. It is obtained by subtracting the initial time from the steady time, i.e. ,like Figure 3 .

[0037] S2, based on the relaxation time of each probe, determine a second frequency for the probe to represent its characteristics; for each probe, apply a second excitation signal including a second DC bias and a sine wave of the second frequency determined for the probe, acquire the corresponding second response signal, and calculate the amplitude of the harmonic components of the second response signal at at least one preset integer multiple of the second frequency.

[0038] The second frequency is the reciprocal of the probe's relaxation time. For example, if the relaxation time of probe A is measured to be 50 μs, the corresponding second frequency is determined to be 20 kHz. If the relaxation time of probe B is measured to be 200 μs, the second frequency is determined to be 5 kHz. A second excitation signal containing, for example, a 150 mV DC bias and a 20 kHz sine wave is applied to probe A. A second excitation signal containing a 150 mV DC bias and a 5 kHz sine wave is applied to probe B. The respective second response signals are acquired, and a Fast Fourier Transform is performed on the signals. The preset integer multiples of the frequency are the second and third harmonics. For probe A, the voltage or current amplitude of the probe response signal at 40 kHz and 60 kHz is calculated. For probe B, the voltage or current amplitude of the probe response signal at 10 kHz and 15 kHz is calculated.

[0039] To tailor a test frequency that reveals the probe's behavior by utilizing its response characteristics, in an optional embodiment, determining a second frequency for representing the characteristics of the probe based on the relaxation time of each probe includes:

[0040] Obtain the relaxation time of each probe. ;

[0041] Calculate the relaxation time The reciprocal of the given value, and the reciprocal value The second frequency is set to the probe.

[0042] Specifically, relaxation time The relaxation time reflects the characteristic rate of charge migration or chemical reaction within the probe contact interface and is the timescale of the process leading to nonlinear effects. The shorter the relaxation time, the faster the interface changes, and the more pronounced the effect will be at high frequencies. The relaxation time of a specific probe is crucial for obtaining this effect. Then, for example, the measurements mentioned above. The second frequency is determined through calculation. This operation involves taking the reciprocal of the relaxation time. The characteristic frequency of the probe can then be calculated. The calculated frequency The reference frequency used as the second excitation signal ensures that the test is performed at a frequency that the probe is sensitive to.

[0043] In an optional embodiment, calculating the harmonic component amplitude of the second response signal at at least a preset integer multiple of the second frequency includes:

[0044] The preset integer multiple frequency is determined as , where k is a preset integer greater than 1;

[0045] Perform a Fast Fourier Transform on the second response signal to obtain the signal spectrum;

[0046] Extract the frequency from the spectrum The signal amplitude at that point is used as the amplitude of the kth harmonic component. .

[0047] Use the second frequency determined in the previous step. For example, at 5 kHz, a second excitation signal is generated and applied to the probe based on a second frequency determined for the probe and its second DC bias. Due to the nonlinear effects of the contact interface, the measured second response signal will contain not only the fundamental frequency. It will also generate signal components with frequencies that are integer multiples of its own, i.e., harmonics. We select harmonics corresponding to integers k greater than 1, such as the second harmonic k=2 and the third harmonic k=3.

[0048] The acquired second response signal is digitized at a sampling frequency of at least 30 kHz and stored as a time-series data block. Subsequently, a windowing function such as the Hanning window is applied to this data block to reduce spectral leakage, and an N-point Fast Fourier Transform (FFT) is performed to convert the signal to the frequency domain, obtaining its complex spectrum. Based on the frequency resolution of the FFT... =Sampling rate / N, and in the calculated spectral amplitude map, precisely locate the frequency index corresponding to the target harmonic frequency. For example, for the second harmonic (10kHz) and third harmonic (15kHz) corresponding to the fundamental frequency of 5kHz, find their precise positions in the spectrum through index calculation, extract the modulus of the complex spectrum at the target frequency, and record them as the amplitude of the second harmonic component. and the amplitude of the third harmonic component See also Figure 4 The magnitude reflects the strength of the nonlinearity of the probe contact.

[0049] S3, for each probe as a target probe, within the probe's spatial neighborhood, at least one reference probe is selected based on the similarity of the fundamental contact resistance between other probes in the neighborhood and the target probe, and a spatial correlation feature vector is constructed for the target probe using the harmonic component amplitudes of the target probe and the reference probe.

[0050] Using probe P0 as the target probe, a radius range is determined based on the x and y coordinates of the target probe on the probe card. All other probes within this range constitute a spatial neighborhood. The percentage difference in fundamental contact resistance between each probe in the neighborhood and P0 is calculated. Probes with a difference percentage less than 5% are selected as reference probes, such as probes P1 and P2. The second harmonic amplitudes of P0, P1, and P2 are then used. , , and third harmonic amplitude Construct the spatial correlation feature vector of P0, for example, the vector can be... and The ratio, and The ratio, and The ratio, and A one-dimensional array consisting of the ratios of .

[0051] To find a set of neighboring probes that are geographically close and have similar basic electrical characteristics for each target probe as a comparison benchmark, in an optional embodiment, the step of selecting at least one reference probe based on the similarity of the fundamental contact resistance of other probes in the neighborhood to the target probe includes:

[0052] Define target probe The three-dimensional coordinates (x, y, z) are given, and a radius is set. A spatial sphere serves as the spatial neighborhood of the probe.

[0053] Calculate every other probe in the neighborhood With target probe The absolute value of the fundamental contact resistance difference ;

[0054] The absolute value of the difference is less than a preset resistance threshold. probe The selected probe is the reference probe.

[0055] Specifically, identify a target probe. The three-dimensional coordinates of the probe on the probe card are obtained. A radius of [missing information] is defined with the probe as the center. A virtual sphere is defined, and all other probes falling within this sphere are considered as candidate probes in the spatial neighborhood.

[0056] Calculate the fundamental contact resistance of the target probe. With each candidate probe in the neighborhood fundamental contact resistance The absolute value of the difference. For example, if the target probe resistance is 5.0 ohms and a neighboring probe resistance is 5.1 ohms, the difference is 0.1 ohms. This difference is then compared to a pre-set resistance threshold. For example, a resistance difference of 0.3 ohms is used for comparison. Only neighboring probes whose absolute resistance difference is less than this threshold are identified as reference probes.

[0057] To construct a feature that can represent the degree of nonlinear anomaly, in an optional embodiment, the step of constructing a spatial correlation feature vector for the target probe using the harmonic component amplitudes of the target probe and the reference probe includes:

[0058] Obtain the amplitude of the kth harmonic component of the target probe and the set of amplitudes of the kth harmonic components of all reference probes , where k is a preset integer;

[0059] Calculate the difference between the amplitude of the k-th harmonic component of the target probe and each reference probe. ;

[0060] All calculated differences The spatial correlation feature vector is formed by arranging the reference probes and target probes according to the spatial distances between the reference probes and target probes corresponding to the differences from near to far.

[0061] Obtain the amplitude of the k-th harmonic component of the target probe, such as the second harmonic amplitude. And the corresponding harmonic amplitudes of all reference probes. Assume the target probe's... The second harmonic amplitude of the target probe is 0.5 mV, and the second harmonic amplitudes of the three reference probes are 0.4 mV, 0.6 mV, and 0.45 mV, respectively.

[0062] The difference in harmonic amplitude between the target probe and each reference probe is calculated one by one. In the example above, the differences obtained are 0.1 mV, -0.1 mV, and 0.05 mV, respectively. These differences reflect the deviation of the nonlinearity of the target probe from its normal neighbors. To include spatial information in the feature vector, the reference probes are sorted from closest to furthest in terms of spatial distance from the target probe. Assuming the distances between the three reference probes and the target probe are 1.5 units, 2.1 units, and 2.8 units, respectively, with corresponding harmonic differences of -0.1 mV, 0.1 mV, and 0.05 mV, the resulting spatial correlation feature vector is [-0.1, 0.1, 0.05].

[0063] S4. Based on the fundamental contact resistance of each probe and the spatial correlation feature vector corresponding to the probe, determine the electrical performance of the semiconductor probe device.

[0064] The fundamental contact resistance value of each probe and all elements of the probe spatial correlation feature vector are concatenated to form a total feature set. This feature set is then input into a pre-trained support vector machine or neural network classification model based on historical data. The model outputs the health status of the probe, such as normal, of interest, or failed. The number and spatial distribution of probes classified as of interest or failed on the entire probe device are counted. If the total number of failed probes exceeds a preset threshold, or if failed probes exhibit spatial clustering, the overall electrical performance of the semiconductor probe device is deemed unqualified.

[0065] The construction and training process of the support vector machine (SVM) or neural network classification model includes: First, collecting historical test data to construct a training set. Each data point corresponds to a probe, and its features are composed of a one-dimensional array formed by concatenating the fundamental contact resistance value of the probe with its spatial correlation feature vector, and labeled with category tags such as "normal," "attention," or "failed" by experts or related test results. For the SVM model, the radial basis function is usually chosen as the kernel function, and its penalty parameter C and kernel width parameter γ are optimized through grid search and cross-validation. For the neural network model, a multilayer perceptron containing several fully connected hidden layers and ReLU activation functions, with the output layer using the Softmax function, and the cross-entropy loss function is used. Before training, the feature data needs to be standardized, such as by Z-score normalization, and the dataset is divided into a training set and a validation set in an approximately 7:3 ratio. The model uses the training set for supervised learning, while monitoring metrics such as accuracy and recall on the validation set to prevent overfitting, using early stopping or adjusting hyperparameters until the model performance stabilizes.

[0066] In an optional embodiment, determining the electrical performance of the semiconductor probe device based on the fundamental contact resistance of each probe and the spatial correlation feature vector corresponding to the probe includes:

[0067] Set the normal range for the fundamental contact resistance. A normal value range is set for each component of the spatial correlation feature vector. ;

[0068] Count the number of probes whose fundamental contact resistance values ​​fall outside the normal range. And the number of probes whose spatial correlation feature vector has at least one component falling outside the normal range. ;

[0069] when Less than the preset maximum number of probes ,and Less than the preset maximum number of probes When the time is right, the electrical performance of the semiconductor probe device is determined to be qualified.

[0070] Specifically, an acceptable range is set for the fundamental contact resistance, such as [4.0 ohms, 6.0 ohms]; and an acceptable fluctuation range is set for each element value in the spatial correlation eigenvector, such as [-0.2 mV, +0.2 mV].

[0071] The total number of probes whose fundamental contact resistance value exceeds the normal range is recorded as follows: Meanwhile, in the statistical spatial correlation feature vector, if any component value exceeds the total number of probes in the normal range, it is denoted as... For example, in a device containing thousands of probes, testing revealed that eight probes had substandard resistance. Five probes showed abnormal spatial correlation, namely The two statistical quantities are compared with a preset upper limit for the number of failed probes, for example... and .because Less than and Less than Since both conditions are met, the overall electrical performance of the semiconductor probe device is deemed qualified.

[0072] In a second embodiment, the present invention also proposes an electrical performance testing system for a semiconductor probe device, comprising the following modules:

[0073] The determination module is used to apply a first excitation signal containing a first DC bias and a first frequency sine wave to multiple probes of the semiconductor probe device, acquire the corresponding first response signal, and calculate the fundamental contact resistance of each probe based on the first response signal; apply a preset step change to the first DC bias, measure the time required for the fundamental contact resistance to recover to a steady state, and determine the relaxation time of each probe.

[0074] The calculation module is used to determine a second frequency for representing the characteristics of each probe based on the relaxation time of each probe; for each probe, apply a second excitation signal including a second DC bias and a sine wave of the second frequency determined for the probe, acquire the corresponding second response signal, and calculate the amplitude of the harmonic components of the second response signal at at least a preset integer multiple of the second frequency.

[0075] The construction module is used to select at least one reference probe in the probe's spatial neighborhood based on the similarity of the fundamental contact resistance of other probes in the neighborhood to the target probe, and to construct a spatial correlation feature vector for the target probe using the harmonic component amplitudes of the target probe and the reference probe.

[0076] The determination module is used to determine the electrical performance of the semiconductor probe device based on the fundamental contact resistance of each probe and the spatial correlation feature vector corresponding to the probe.

[0077] In an optional embodiment, calculating the fundamental contact resistance of each probe based on the first response signal includes:

[0078] The first frequency sine wave in the first excitation signal is used as the input current signal I(t), and the AC voltage signal V(t) in the first response signal is collected.

[0079] The amplitudes of the input current signal and the acquired AC voltage signal at the first frequency were extracted using Fourier transform. and ;

[0080] According to Ohm's law, the magnitude of the fundamental contact resistance is calculated. .

[0081] In an optional embodiment, the measurement of the time required for the fundamental contact resistance to recover to a steady state, and the determination of the relaxation time of each probe, include:

[0082] At any moment Apply a DC bias to the first DC bias by to The step voltage;

[0083] From time Initially, the fundamental contact resistance sequence was continuously measured and recorded at a fixed sampling period T. ;

[0084] Calculate the absolute value of the difference between adjacent measurements in the sequence. When the absolute value is less than a preset steady-state criterion threshold for a preset number of consecutive times At this time, record the current time as ;

[0085] The relaxation time Determined as .

[0086] In an optional embodiment, determining a second frequency for representing the characteristics of the probe based on the relaxation time of each probe includes:

[0087] Obtain the relaxation time of each probe. ;

[0088] Calculate the relaxation time The reciprocal of the given value, and the reciprocal value The second frequency is set to the probe.

[0089] In an optional embodiment, calculating the harmonic component amplitude of the second response signal at at least a preset integer multiple of the second frequency includes:

[0090] The preset integer multiple frequency is determined as , where k is a preset integer greater than 1;

[0091] Perform a Fast Fourier Transform on the second response signal to obtain the signal spectrum;

[0092] Extract the frequency from the spectrum The signal amplitude at that point is used as the amplitude of the kth harmonic component. .

[0093] In an optional embodiment, the step of selecting at least one reference probe based on the fundamental contact resistance similarity between other probes in the neighborhood and the target probe includes:

[0094] Define target probe The three-dimensional coordinates (x, y, z) are given, and a radius is set. A spatial sphere serves as the spatial neighborhood of the probe.

[0095] Calculate every other probe in the neighborhood With target probe The absolute value of the fundamental contact resistance difference ;

[0096] The absolute value of the difference is less than a preset resistance threshold. probe The selected probe is the reference probe.

[0097] In an optional embodiment, constructing a spatial correlation feature vector for the target probe using the harmonic component amplitudes of the target probe and the reference probe includes:

[0098] Obtain the amplitude of the kth harmonic component of the target probe and the set of amplitudes of the kth harmonic components of all reference probes , where k is a preset integer;

[0099] Calculate the difference between the amplitude of the k-th harmonic component of the target probe and each reference probe. ;

[0100] All calculated differences The spatial correlation feature vector is formed by arranging the reference probes and target probes according to the spatial distances between the reference probes and target probes corresponding to the differences from near to far.

[0101] In an optional embodiment, determining the electrical performance of the semiconductor probe device based on the fundamental contact resistance of each probe and the spatial correlation feature vector corresponding to the probe includes:

[0102] Set the normal range for the fundamental contact resistance. A normal value range is set for each component of the spatial correlation feature vector. ;

[0103] Count the number of probes whose fundamental contact resistance values ​​fall outside the normal range. And the number of probes whose spatial correlation feature vector has at least one component falling outside the normal range. ;

[0104] when Less than the preset maximum number of probes ,and Less than the preset maximum number of probes When the time is right, the electrical performance of the semiconductor probe device is determined to be qualified.

[0105] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0106] The functional modules shown in the above-described block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0107] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0108] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0109] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for testing the electrical performance of a semiconductor probe device, characterized in that, include: A first excitation signal comprising a first DC bias and a first frequency sine wave is applied to multiple probes of the semiconductor probe device, a corresponding first response signal is acquired, and the fundamental contact resistance of each probe is calculated based on the first response signal, wherein the first frequency is the AC component of the first excitation signal; a preset step change is applied to the first DC bias, and the time required for the fundamental contact resistance to recover to a steady state is measured to determine the relaxation time of each probe. The step of calculating the fundamental contact resistance of each probe based on the first response signal includes: The first frequency sine wave in the first excitation signal is used as the input current signal I(t), and the AC voltage signal V(t) in the first response signal is acquired; the amplitudes of the input current signal and the acquired AC voltage signal at the first frequency are extracted using Fourier transform. and According to Ohm's law, the magnitude of the fundamental contact resistance was calculated. ; Based on the relaxation time of each probe, a second frequency for representing the characteristics of the probe is determined, specifically by: acquiring the relaxation time of each probe. ; Calculate the relaxation time The reciprocal of the given value, and the reciprocal value The second frequency of the probe is set; for each probe, a second excitation signal including a second DC bias and a sine wave of the second frequency determined for the probe is applied, the corresponding second response signal is acquired, and the amplitude of the harmonic components of the second response signal at at least a preset integer multiple of the second frequency is calculated. For each probe as a target probe, within the probe's spatial neighborhood, at least one reference probe is selected based on the similarity of the fundamental contact resistance between other probes in the neighborhood and the target probe. Then, using the harmonic component amplitudes of the target probe and the reference probe, a spatial correlation feature vector is constructed for the target probe. Specifically, this includes obtaining the k-th harmonic component amplitude of the target probe. and the set of amplitudes of the kth harmonic components of all reference probes , where k is a preset integer; calculate the difference between the amplitude of the kth harmonic component of the target probe and each reference probe; combine all the calculated differences The spatial correlation feature vector is formed by arranging the reference probes and target probes corresponding to the differences from near to far. Based on the fundamental contact resistance of each probe and the spatial correlation feature vector corresponding to the probe, the electrical performance of the semiconductor probe device is determined. Specifically, this includes: concatenating the fundamental contact resistance value of each probe and all elements of the probe spatial correlation feature vector into a total feature set; inputting the total feature set into a support vector machine or neural network classification model pre-trained with historical data, and outputting the health status of the probes, including normal, attention, and failure status; counting the number and spatial distribution of probes determined to be in attention or failure status on the entire probe device; if the total number of probes in failure status exceeds a preset threshold, or if probes in failure status exhibit spatial clustering, then the overall electrical performance of the semiconductor probe device is determined to be unqualified.

2. The method according to claim 1, characterized in that, The measurement of the time required for the fundamental contact resistance to recover to a steady state, and the determination of the relaxation time of each probe, include: At any moment Apply a DC bias to the first DC bias by to The step voltage; From time Initially, the fundamental contact resistance sequence was continuously measured and recorded at a fixed sampling period T. ; Calculate the absolute value of the difference between adjacent measurements in the sequence. When the absolute value is less than a preset steady-state criterion threshold for a preset number of consecutive times... At this time, record the current time as ; The relaxation time Determined as .

3. The method according to claim 1, characterized in that, The calculation of the harmonic component amplitude of the second response signal at at least one preset integer multiple of the second frequency includes: The preset integer multiple frequency is determined as , where k is a preset integer greater than 1; Perform a Fast Fourier Transform on the second response signal to obtain the signal spectrum; Extract the frequency from the spectrum The signal amplitude at that point is used as the amplitude of the kth harmonic component. .

4. The method according to claim 1, characterized in that, The step of selecting at least one reference probe based on the similarity of the fundamental contact resistance of other probes in the neighborhood to that of the target probe includes: Define target probe The three-dimensional coordinates are given, and a radius is set. A spatial sphere serves as the spatial neighborhood of the probe. Calculate every other probe in the neighborhood With target probe The absolute value of the fundamental contact resistance difference; The absolute value of the difference is less than a preset resistance threshold. probe The selected probe is the reference probe.

5. The method according to claim 1, characterized in that, The method of determining the electrical performance of the semiconductor probe device based on the fundamental contact resistance of each probe and the spatial correlation feature vector corresponding to the probe further includes: Set the normal range for the fundamental contact resistance. A normal value range is set for each component of the spatial correlation feature vector. ; Count the number of probes whose fundamental contact resistance values ​​fall outside the normal range. And the number of probes whose spatial correlation feature vector has at least one component falling outside the normal range. ; when Less than the preset maximum number of probes ,and Less than the preset maximum number of probes When the time is right, the electrical performance of the semiconductor probe device is determined to be qualified.

6. An electrical performance testing system for a semiconductor probe device, characterized in that, Includes the following modules: A determination module is configured to apply a first excitation signal comprising a first DC bias and a first frequency sine wave to multiple probes of the semiconductor probe device, acquire a corresponding first response signal, and calculate the fundamental contact resistance of each probe based on the first response signal, wherein the first frequency is the AC component of the first excitation signal; apply a preset step change to the first DC bias, measure the time required for the fundamental contact resistance to recover to a steady state, and determine the relaxation time of each probe; wherein, the calculation of the fundamental contact resistance of each probe based on the first response signal includes: using the first frequency sine wave in the first excitation signal as an input current signal I(t), and acquiring the AC voltage signal V(t) in the first response signal; using Fourier transform to extract the sum of the amplitudes of the input current signal and the acquired AC voltage signal at the first frequency; and calculating the magnitude of the fundamental contact resistance according to Ohm's law; The calculation module is used to determine a second frequency for representing the characteristics of each probe based on the relaxation time of each probe, specifically by acquiring the relaxation time of each probe. ; Calculate the relaxation time The reciprocal of the given value, and the reciprocal value The second frequency of the probe is set; for each probe, a second excitation signal including a second DC bias and a sine wave of the second frequency determined for the probe is applied, the corresponding second response signal is acquired, and the amplitude of the harmonic components of the second response signal at at least a preset integer multiple of the second frequency is calculated. The construction module is used to treat each probe as a target probe, and within the probe's spatial neighborhood, to select at least one reference probe based on the similarity of the fundamental contact resistance of other probes in the neighborhood to the target probe. It then uses the harmonic component amplitudes of the target probe and the reference probe to construct a spatial correlation feature vector for the target probe. Specifically, this includes obtaining the k-th harmonic component amplitude of the target probe. and the set of amplitudes of the kth harmonic components of all reference probes , where k is a preset integer; calculate the difference between the amplitude of the kth harmonic component of the target probe and each reference probe; combine all the calculated differences The spatial correlation feature vector is formed by arranging the reference probes and target probes corresponding to the differences from near to far. The determination module is used to determine the electrical performance of the semiconductor probe device based on the fundamental contact resistance of each probe and the spatial correlation feature vector corresponding to the probe. Specifically, it includes: concatenating the fundamental contact resistance value of each probe and all elements of the probe spatial correlation feature vector into a total feature set; inputting the total feature set into a support vector machine or neural network classification model pre-trained with historical data, and outputting the health status of the probes, including normal, attention, and failure status; counting the number and spatial distribution of probes determined to be in attention or failure status on the entire probe device; if the total number of failure probes exceeds a preset threshold, or if failure probes exhibit spatial clustering, the overall electrical performance of the semiconductor probe device is determined to be unqualified.

Images