Method for evaluating surge current resistance of silicon carbide diode

Abstract

The invention discloses a method for evaluating the surge current resistance of a silicon carbide diode. The method comprises the following steps: acquiring reverse I-V characteristic data of a device at different temperatures in a wafer test stage; according to the technology, a Frechet distance algorithm is introduced according to the positive correlation between the change rate of leakage current along with temperature and the surge peak current of a device. In the evaluation process, a judgment threshold value Dthresh is set according to the rated surge current, the sum Dtotal of Freet distances is calculated, and the surge current resistance of the device can be accurately evaluated by comparing the Dtotal with the Dthresh. As a nondestructive evaluation method, the device with high reliability can be effectively screened out in the wafer test stage, a solution is provided for the reliability of the device, and therefore the guarantee level of product quality is improved.

Description

Technical Field

[0001] The present invention relates to a method for evaluating the surge current resistance capability of a silicon carbide diode, and belongs to the field of reliability evaluation of silicon carbide semiconductor devices. Background Art

[0002] As a wide bandgap semiconductor device, silicon carbide (SiC) diodes have gradually become important components in the field of power electronics due to their excellent high temperature, high voltage and high frequency characteristics. However, in actual applications, silicon carbide diodes face many potential surge shocks, and these surges will have a significant impact on the long-term stability and reliability of the device. Therefore, how to effectively evaluate the surge capability of silicon carbide diodes has become a key issue to ensure their application reliability and lifespan. For SiC diodes, the impact of surges may not only cause thermal damage to the internal materials of the device, but may also cause breakdown of the diode device or thermal failure of the structure, thereby directly affecting its performance. Therefore, the assessment of surge withstand capability has become an important indicator to measure whether SiC diodes can operate stably in a working environment. The evaluation method proposed in the present invention combines variable temperature testing technology with surge reliability assessment, breaks through the limitations of traditional testing methods, and not only provides a new device surge withstand capability assessment method, but also provides a more convenient and scientific means for quality control and reliability verification of silicon carbide devices. Summary of the Invention

[0003] Purpose of the invention: In view of the shortcomings of existing technologies, the purpose of the present invention is to provide a method for evaluating the surge current resistance of silicon carbide diodes that can efficiently cope with the rapid screening of a large number of devices and reduce the complexity of evaluation.

[0004] Technical solution: The method for evaluating the surge current resistance capability of a silicon carbide diode described in the present invention comprises the following steps:

[0005] (A1) Apply different voltages to the SiC diode at different temperatures and record the reverse current of the SiC diode at different voltages;

[0006] (A2) Based on the reverse current and voltage data, reverse IV characteristic curves at different temperatures are summarized;

[0007] (A3) using the reverse IV characteristic curve of the silicon carbide diode at the wave stabilization temperature obtained in step (A2) as a reference curve, and using a Frechet distance algorithm to calculate the Frechet distances between the reverse IV characteristic curves at other different temperatures and the reference curve;

[0008] (A4) Calculate the sum of the Frechet distances D under each temperature condition total ;

[0009] (A5) Setting a threshold D_thresh for evaluating the surge current capability of a silicon carbide diode based on the Frechet distance;

[0010] (A6) By comparing D total Evaluating the surge reliability of SiC diodes with D_thresh.

[0011] Furthermore, in step (A1), the different temperature ranges include low temperature, room temperature and high temperature.

[0012] Furthermore, the temperature range of the different temperatures is 25-200°C.

[0013] Furthermore, in step (A1), the number of temperature points set at different temperatures is more than six. The testing process at different temperatures does not require disassembly of the package, thereby ensuring testing efficiency and avoiding possible secondary damage.

[0014] Furthermore, in step (A2), the reverse IV characteristic curves at different temperatures are summarized using Origin drawing software.

[0015] Furthermore, in step (A3), the Frechet distance is calculated using the following formula (1):

[0016]

[0017] In formula (1), d Frechet (C base , C i ) describes two curves C base and C i The Frechet distance of the curve is expressed as the minimum of the maximum distances between corresponding points under all possible path parameterizations. Specifically, d Frech (C base , C i ) is the curve C at other temperatures i With the reference curve C base The Frechet distance between Reference curve C base The path parameterization function on the curve C at other temperatures is i parameterized function on the paths, allowing them to advance at different speeds, while is the Euclidean distance between corresponding points, means to do everything possible and ψ functions to find an infimum, means that among all possible reparameterized functions and ψ(t), curve C iWith the reference curve C base The maximum value of the corresponding point with the smallest distance from the maximum value is the value of the two curves C base and C i The Fréchet distance between the two curves measures the overall closeness of the two curves under various reasonable simultaneous ergodic comparisons.

[0018] Furthermore, in step (A3), the reverse IV characteristic curve at room temperature (25° C.) is selected as the reference curve.

[0019] Furthermore, in step (A4), the sum of the Frechet distances D is calculated. total The following formula (2) is used for calculation:

[0020]

[0021] In formula (2), D total is the sum of the Frechet distances of each curve, C base Indicated as the set reference curve, C i is the reverse IV characteristic curve at other temperatures, d Frechet (C base , C i ) is C base and the Fréchet distance of Ci.

[0022] Furthermore, in step (A5), setting a threshold D_thresh, an evaluation standard for the surge current resistance capability of the silicon carbide diode, based on the Frechet distance includes the following steps:

[0023] The threshold D_thresh is related to the rated current of the SiC diode. For SiC diodes with higher rated currents, a higher threshold is set to ensure the diode's surge current resistance can accommodate higher loads. For SiC diodes with lower rated currents, a lower threshold is set. Generally, steps (A1)-(A4) are repeated using SiC diodes capable of withstanding surges of 2-10 times the rated current, depending on the needs. The resulting D_total is used as the evaluation standard threshold D_thresh.

[0024] Further, in step (A6), according to D total Compare the result with D_thresh to judge whether the silicon carbide diode has good surge reliability. total If D_thresh is greater than D_thresh, the silicon carbide diode is judged to have strong surge withstand capability; otherwise, if D total If it is less than D_thresh, it is determined that the silicon carbide diode has weak surge current resistance capability.

[0025] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages:

[0026] The present invention proposes a method for evaluating the surge reliability of silicon carbide diodes based on variable temperature testing, which can efficiently and accurately evaluate the surge capability of the device. By testing the reverse IV characteristics under a variable temperature environment and analyzing the characteristic curves at different temperatures, the Frechet distance is combined to quantitatively evaluate the similarity of the curves, and then the surge reliability of the device is judged. This not only simplifies the testing process, but also improves the accuracy and efficiency of the evaluation. The present invention provides an efficient detection method based on temperature changes. This method breaks through the traditional testing method, is suitable for large-scale production and quality control, can effectively improve the prediction accuracy of surge withstand capability, and has significant technical advantages and application value. As a non-destructive evaluation method, the present invention can also screen out high-reliability devices with strong surge current resistance in real time during the wafer testing stage, providing an important basis for subsequent reliability testing of the device. This method can not only avoid the use of low-reliability devices in the early stages, but also greatly improve the level of product quality assurance, and has significant application value.

[0027] The present invention collects reverse IV characteristic data of the device at different temperatures during the wafer testing stage. This technology introduces the Frechet distance algorithm based on the positive correlation between the rate of change of leakage current with temperature and the peak surge current of the device. In the evaluation process, the judgment threshold D_thresh is set according to the rated surge current, and the sum of the Frechet distances D_total is calculated. By comparing D_total with D_thresh, the surge current resistance capability of the device can be accurately evaluated. As a non-destructive evaluation method, the present invention can effectively screen out devices with high reliability during the wafer testing stage, providing a solution for the reliability of the device, thereby improving the level of product quality assurance. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a schematic flow chart of the evaluation method of the present invention;

[0029] Figure 2 is a reverse IV characteristic curve diagram of device A and device B under variable temperature conditions in Example 1;

[0030] Figure 3 is a graph showing the sum of the Frechet distances and the Frechet distances between the reverse IV characteristic curves in Example 1;

[0031] Figure 4 1 is a diagram showing the surge reliability evaluation results of the device in the embodiment of the present invention; DETAILED DESCRIPTION

[0032] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0033] Example 1

[0034] like Figure 1 As shown, the present invention designs a method for evaluating the surge current resistance of silicon carbide diodes. By using a high-temperature environmental test chamber and measuring the reverse IV characteristics at different temperatures, the performance stability of the diode under various operating environments is comprehensively evaluated. First, the electrical data of the diode is obtained by performing reverse IV characteristic tests at room temperature, low temperature, and high temperature. Then, based on the test results at different temperatures, the Frechet distance algorithm is used to quantitatively analyze the similarity of the reverse IV curves to determine the surge resistance of the device. The surge reliability of the device is evaluated by setting the sum of the Frechet distances between the baseline curve and other temperature curves and comparing it with the threshold corresponding to the rated current of the device.

[0035] The specific implementation is as follows:

[0036] Step (1): Setting up a high-temperature environmental test box and placing the diode: First, you need to configure an environmental test box with adjustable temperature. The environmental test box used can accurately adjust the ambient temperature, covering a variety of temperature conditions such as normal temperature and high temperature (the temperature adjustment range is 25-200°C). The design of the test box takes into account the thermal cycle at different temperatures, simulating the temperature fluctuations and thermal stress experienced by silicon carbide diodes in real applications. In order to comprehensively evaluate the performance of the diode, six test temperatures are set in this embodiment, namely: normal temperature (25°C), 60°C, 90°C, 120°C, 150°C and 200°C. These temperatures will cover a wide range from normal temperature to high temperature, testing the stability and reliability of the diode under different thermal stresses.

[0037] After the parameters of the high-temperature environmental test chamber are set, the diodes to be tested are placed in the high-temperature environmental test chamber one by one. In this embodiment, two silicon carbide diodes with a rated current of 40A are selected and named as device A and device B respectively.

[0038] Step (2): Variable temperature test of reverse IV characteristics of silicon carbide diode devices: In order to fully evaluate the performance of silicon carbide diodes, it is necessary to test their reverse IV characteristics under different temperature conditions. During the test, first apply a voltage of 0 to 1800V to the cathode of the device, record the corresponding reverse current under this voltage, and adjust the temperature through the temperature control system, gradually changing the ambient temperature from room temperature to high temperature. After each temperature change, the device needs to be kept at this temperature for fifteen minutes to stabilize the performance of the device before the reverse IV characteristics test is carried out to ensure that its electrical characteristics are not affected by instantaneous changes.

[0039] Based on the applied voltage and the corresponding recorded reverse current, a reverse IV curve at the current temperature was obtained using Origin plotting software. Finally, six reverse IV curves were measured for device A and device B at 25°C, 60°C, 90°C, 120°C, 150°C, and 200°C, respectively. At this point, these curves were still separate. To better evaluate the device's surge capability, the reverse IV curves under varying temperatures needed to be integrated into a single graph. The measured reverse IV characteristic data can be used to obtain the reverse current response curves of the diode at different temperatures. These data are key to evaluating the device's electrical characteristics and stability.

[0040] Step (3): Data processing and summary. Integrate the reverse current-voltage curves at different temperatures to provide a data basis for subsequent analysis. Summarize the reverse IV characteristic curves of device A and device B. Each figure contains the reverse IV characteristic curves of a device at all temperatures (25°C, 60°C, 90°C, 120°C, 150°C and 200°C). Finally, a summary of the temperature-varying reverse IV characteristic curves of the two devices is obtained. Figure 2 shown. Figure 2 This is the reverse IV characteristic curve of device A and device B under variable temperature conditions in Example 1, where a is device A and b is device B. Figure 2 The degree of fit of the middle curve is used to evaluate the surge resistance of the device.

[0041] Step (4): Based on the rated current of the device, set the threshold value D_thresh of the sum of the Frechet distances between the curves. This threshold value is the sum of the Frechet distances between the curves measured under the same temperature conditions for devices with strong surge reliability. total Comparison is used to evaluate the surge current capability of silicon carbide diodes.

[0042] Silicon carbide diodes have strong high-temperature and high-power carrying capabilities, can withstand surge currents several times the rated current, and exhibit strong surge withstand capabilities. In this embodiment, a device capable of withstanding 210A (seven times the rated current) is selected. The sum of the Frechet distances D_total between the IV characteristic curves measured at different temperatures (25°C, 60°C, 90°C, 120°C, 150°C, and 200°C) is used as the threshold D_thresh for evaluating the surge current resistance of the silicon carbide diode. D_total in this case is calculated, and the D_total value of the device at the six test temperatures is obtained as 668.418. In this embodiment, the value of the threshold D_thresh is defined as 668.418.

[0043] When evaluating the surge withstand capability of devices A and B, the D_total of devices A and B at six test temperatures are compared with the set threshold D_thresh.

[0044] The device can withstand surges of seven times its rated current, meaning it can withstand extremely high current loads for a short period of time without damage, enabling it to demonstrate excellent performance and reliability in high-power, high-temperature, and high-frequency applications. This capability is crucial for ensuring the stable operation of power electronic equipment in harsh environments.

[0045] Step (5): In step (3), the degree of curve fit is evaluated by calculating the Frechet distance, which is calculated using the following formula (1):

[0046]

[0047] In formula (1), d Frechet (C base , C i ) describes two curves C base and C i The Frechet distance of the curve is expressed as the minimum of the maximum distances between corresponding points under all possible path parameterizations. Specifically, d Frech (C base , C i ) is the curve C at other temperatures i With the reference curve C base The Frechet distance between Reference curve C base The path parameterization function on the curve C at other temperatures is i parameterized function on the paths, allowing them to advance at different speeds, while is the Euclidean distance between corresponding points. means to do everything possible and ψ functions to find an infimum, means that among all possible reparameterized functions and ψ(t), find the curve C i With the reference curve C base The maximum value of the corresponding point with the smallest distance from the maximum value is the value of the two curves C base and C i The Fréchet distance between the two curves measures the overall closeness of the two curves under various reasonable simultaneous ergodic comparisons.

[0048] The sum of the Frechet distances between the curves D_total is calculated by the following formula (2):

[0049]

[0050] In the above formula (2), D total is the sum of the Frechet distances of each curve, C base Indicated as the set reference curve, C i is the reverse IV characteristic curve at other temperatures, d Frechet It is C base and the Fréchet distance of Ci.

[0051] In this embodiment, the reverse IV characteristic curve at room temperature (25°C) is selected as the reference curve C base , C i Reverse IV curves at other temperatures (60℃, 90℃, 120℃, 150℃ and 200℃). Solve for the total D total As a metric, it will be compared with a set threshold value D_thresh, and the comparison result will be used to evaluate the difference or similarity between different IV curves, and then used to estimate the surge reliability of SiC diodes.

[0052] In MATLAB, you can calculate the Fréchet distance between two curves by writing a dynamic programming algorithm. First, define the coordinate data of the two curves (such as curve1 and curve2). Next, write the frechet_distance function, which uses dynamic programming to calculate the distance between the curves point by point and update the dp table to ultimately obtain the Fréchet distance between the two curves. Then, use this function to calculate the distance between curve1 and other curves (such as curve2, curve3, etc.). Finally, output the Fréchet distance between each pair of curves and the reference curve, as well as their sum. The dynamic programming method ensures efficient calculation by storing intermediate results.

[0053] The reverse IV characteristic curve at room temperature (25°C) is selected as the reference curve (C base ), then calculate the Frechet distance between the reference curve and the reverse IV curves at other temperatures (60°C, 90°C, 120°C, 150°C, and 200°C), and calculate the total Frechet distance D_total between the multiple reverse IV curves and the reference curve. Use MTLAB programming software and Frechet distance to complete the above calculations. The results are as follows: Figure 3 shown. Figure 3is a graph showing the sum of the Frechet distances between the reverse IV characteristic curves in Example 1, wherein C_base-C_2 is the Frechet distance between the 60°C reverse IV curve and the reference curve, C_base-C_3 is the Frechet distance between the 90°C reverse IV curve and the reference curve, C_base-C_4 is the Frechet distance between the 120°C reverse IV curve and the reference curve, C_base-C_5 is the Frechet distance between the 150°C reverse IV curve and the reference curve, and C_base-C_6 is the Frechet distance between the 200°C reverse IV curve and the reference curve; and D_total is the sum of the Frechet distances between the curves.

[0054] In this embodiment, the Frechet distances between the reference curve of device A and the reverse IV curves at other temperatures are 5.373 (60°C), 41160 (90°C), 78.000 (120°C), 113.69 (150°C), and 231.288 (200°C), respectively. The total Frechet distances D_total between the multiple reverse IV curves and the reference curve is 468.99. The Frechet distances between the reference curve of device B and the reverse IV curves at other temperatures are 40.545 (60°C), 79.620 (90°C), 127.695 (120°C), 191.415 (150°C), and 419.061 (200°C), respectively. The D_total of device B is 858.336.

[0055] Step (6): Compare D_total = 468.99 of device A and D_total = 858.336 of device B with the set threshold D_thresh = 668.418. If D_total is greater than D_thresh, it means that the diode can maintain good performance under different temperatures and the surge reliability of the device is strong; conversely, if D_total is less than D_thresh, it indicates that the performance of the device under temperature changes is poor and the surge reliability is weak.

[0056] The D_total of device A is 468.99, which is less than the set threshold D_thresh, while the D_total of device B is 858.336, which is greater than the set threshold D_thresh. Figure 4 shown. Figure 4This figure shows the surge reliability assessment results for devices A and B in Example 1. Device B's surge withstand capability is higher than seven times the rated current, indicating strong surge withstand capability. Device A's surge withstand capability does not reach seven times the rated current, indicating weaker surge withstand capability. In variable temperature environments, device A's surge withstand capability is inferior to that of device B. This method allows for a dynamic assessment of the surge withstand capability of silicon carbide diodes, taking into account both the device's rated current and actual performance, providing a scientific basis for device reliability and application.

Claims

1. A method for evaluating the surge current resistance capability of a silicon carbide diode, characterized in that: The following steps are involved: (A1) Apply different voltages to the silicon carbide diode at different temperatures and record the reverse current of the silicon carbide diode at different voltages; (A2) Based on the reverse current and voltage data, reverse IV characteristic curves at different temperatures are summarized; (A3) The reverse IV characteristic curve of the silicon carbide diode at the stable temperature obtained in step (A2) is used as the reference curve C base , use the Frechet distance algorithm to calculate the reverse IV characteristic curves at other different temperatures and the reference curve C base Frechet distance; (A4) Calculate the sum of the Frechet distances D under each temperature condition total ; (A5) Setting a threshold D_thresh for evaluating the surge current capability of a silicon carbide diode based on the Frechet distance; (A6) By comparing D total Evaluating the surge reliability of SiC diodes with D_thresh.

2. The evaluation method according to claim 1, characterized in that: In step (A1), the different temperature ranges include low temperature, normal temperature and high temperature.

3. The evaluation method according to claim 2, characterized in that: In step (A1), the temperature range of the different temperatures is 25-200°C.

4. The evaluation method according to claim 1, characterized in that: In step (A1), the number of temperature points set at different temperatures is more than six points.

5. The evaluation method according to claim 1, characterized in that: In step (A2), the characteristics are summarized into a characteristic curve using Origin drawing software.

6. The evaluation method according to claim 1, characterized in that: In step (A3), the Frechet distance is calculated using the following formula (1): In formula (1), d Frechet (C base ,C i ) is the curve C at other temperatures i With reference curve C base The Frechet distance between Reference curve C base The path parameterization function on the curve C at other temperatures is i Path parameterized function on , is the Euclidean distance between corresponding points, means that among all possible reparameterized functions and ψ(t), curve C i With reference curve C base The maximum value whose corresponding point is the smallest distance from the maximum value.

7. The evaluation method according to claim 1, characterized in that: In step (A3), the reverse IV characteristic curve at room temperature is selected as the reference curve C base .

8. The evaluation method according to claim 1, characterized in that: In step (A4), the sum of the Frechet distances D is calculated. total The following formula (2) is used for calculation: In formula (2), D total is the sum of the Frechet distances of each curve, C base is the set reference curve, C i is the reverse IV characteristic curve at other temperatures, d Frechet (C base ,C i ) is C base The Frechet distance between θ and Ci.

9. The evaluation method according to claim 1, characterized in that: In step (A5), the evaluation standard D_thresh of the surge current resistance capability of the silicon carbide diode is set based on the Frechet distance, including the following steps: Repeat steps (A1)-(A4) using a silicon carbide diode that can withstand 2-10 times the rated current surge as needed. total As the evaluation criterion threshold D_thresh.

10. The evaluation method according to claim 1, characterized in that: In step (A6), according to D total With D thres The comparison results can be used to determine whether the silicon carbide diode has good surge reliability. total Greater than D thres , it is determined that the silicon carbide diode has a strong surge withstand capability; on the contrary, if D total Less than D thresh , it is determined that the silicon carbide diode has weak surge current resistance capability.

Images