Thyristor failure test device

Abstract

The invention discloses a thyristor failure test device, which relates to the technical field of thyristor life test, and comprises a power supply, a protective resistor, an artificial neutron source, a data acquisition device and a processing module, and is characterized in that the power supply is used for providing direct current bias voltages with different sizes and different polarities for a sample thyristor; the artificial neutron source is used for enabling random single-particle burnout events to occur in the sample thyristor; the protection resistor is used for limiting main loop current; the data acquisition equipment is used for acquiring bias voltage signals on the sample thyristors in real time; and the processing module is used for recording the occurrence time of the single-particle burnout event according to the bias voltage signal on the sample thyristor when the single-particle burnout event occurs in the sample thyristor, and fitting the occurrence time of the single-particle burnout event and the corresponding bias voltage signal to obtain the failure rate of the thyristor. According to the invention, the condition of irradiation failure of the thyristor under voltage bias under the actual working condition can be truly reflected.

Description

Technical Field

[0001] The invention relates to the technical field of neutron irradiation thyristor effect test, in particular to a thyristor failure test device. Background Art

[0002] At present, my country's DC transmission technology is developing rapidly and is in an international leading position. With the proposal of "exporting electricity from Tibet", there will be more and more converter stations in high-altitude areas of my country. The higher the altitude of the converter station, the higher the atmospheric neutron flux, the higher the failure risk faced by the thyristor, the core component of the converter valve, and the higher the failure rate, which may threaten the safe and stable operation of the high-voltage DC transmission system.

[0003] For thyristors, the main reason for irradiation failure under working conditions is single particle burnout, that is, high-energy neutrons are incident on the thyristor to generate space charge. Under the action of external voltage, a large number of carriers are generated in the local area of ​​the device, the temperature rises, and thermal breakdown and burnout occur inside the device. Therefore, there are two possible reasons for the failure of thyristors due to atmospheric neutron irradiation: one is that atmospheric neutron irradiation directly acts on the thyristor to cause single particle effect, especially single particle burnout events leading to device failure; the other is that atmospheric neutrons cause failures in the control part of the high-voltage direct current transmission system, thereby indirectly causing thyristor failure.

[0004] Lifetime parameters are a very important component of reliability parameters, and there is a large demand for lifetime assessment in newly developed equipment and in-service equipment. For the threat of thyristor failure caused by atmospheric neutron irradiation, accelerated life failure tests can be performed based on artificial neutron sources to obtain the failure rate of thyristors.

[0005] Existing failure tests mainly include the 1080.1 experiment in the US military standard MIL-STD-750, the mechanical test of semiconductor devices in the national standard GB / T4937.17-2018, and the radiation hardening test of semiconductor devices in the national military standard GJB 762.1-89.

[0006] The experiment 1080.1 in the US military standard MIL-STD-750 briefly describes the irradiation test specifications for single-event burnout (SEB) and single-event gate rupture (SEGR) of planar vertical power MOSFETs under heavy ion irradiation. However, thyristors do not have the oxide layer structure of MOSFET-type devices, and only SEB events will occur after neutron irradiation, so the experimental methods in this standard specification cannot be directly applied to thyristors.

[0007] The national military standard GJB 762.1-89 regulates the radiation hardening test of semiconductor devices. However, this method is mainly proposed for MOSFET type devices, and some settings in the method (such as pin short circuit) are not applicable to the experimental design of thyristors. The standard stipulates that the irradiation source used is a fast neutron reactor, a TRIGA reactor, a steady-state reactor or a 14MeV neutron generator, and is not suitable for accelerated life tests under artificial neutron source environments. The devices to be tested in this standard are in a non-biased state during irradiation, and the main research content is the impact of single irradiation stress on the life of the device, which is difficult to reflect the radiation damage of thyristors under voltage bias under actual working conditions. The standard requires no less than 10 experimental devices in the same batch, which is cost-unacceptable for high-voltage and high-current discrete devices such as thyristors.

[0008] The national standard GB / T 4937.17-2018 is mainly aimed at MOSFET devices and is not suitable for the experimental design of thyristors. The standard lacks definition of key parameters for judging device degradation. The radiation source defined in the standard is a pulse reactor, and the final evaluation method is the equivalent injection of 1MeV. Its energy spectrum is too different from that of atmospheric neutrons, which is not suitable for the application scenario of accelerated equivalence research for atmospheric neutron irradiation.

[0009] In summary, existing failure tests mainly focus on metal-oxide semiconductor field-effect transistors (MOSFETs). The large number of devices required to be tested leads to excessively high costs. In addition, the devices to be tested are in a non-biased state during irradiation, which makes it difficult to reflect the thyristor irradiation failure under voltage bias under actual working conditions. Summary of the invention

[0010] The present invention provides a thyristor failure test device, which solves the problems that the existing failure test requires a large number of devices to be tested, resulting in high costs, and the devices to be tested are in a non-biased state during irradiation, making it difficult to reflect the thyristor radiation damage under voltage bias under actual working conditions.

[0011] The present invention provides a thyristor failure test device, comprising:

[0012] A power supply, used for providing DC bias voltages of different magnitudes and polarities to the sample thyristors;

[0013] An artificial neutron source is used to generate neutron beams with different irradiation dose rates, and irradiate the sample thyristor with the neutron beam to cause a random single particle burning event inside the sample thyristor;

[0014] A protective resistor, one end of which is connected to the output end of the power supply, and the other end is connected to one end of the sample thyristor. When a single-particle burnout event occurs in the sample thyristor, a conductive channel is formed inside the sample thyristor, causing the main circuit current to rise. At this time, the protective resistor is used to limit the main circuit current;

[0015] A data acquisition device is electrically connected to the sample thyristor and is used to collect bias voltage signals on the sample thyristor in real time;

[0016] The processing module is used to record the occurrence time of the single-particle burnout event according to the bias voltage signal on the sample thyristor when a single-particle burnout event occurs in the sample thyristor under DC bias voltages of different sizes and polarities, and to perform fitting processing on multiple occurrence times of the single-particle burnout event and the corresponding bias voltage signals to obtain the thyristor failure rate.

[0017] Preferably, the sample thyristor is arranged in a neutron beam channel of an artificial neutron source.

[0018] Preferably, the data acquisition device includes a voltage probe and an oscilloscope.

[0019] Preferably, a thermocouple is provided on the outside of the sample thyristor, and the thermocouple is used to measure the temperature of the sample thyristor.

[0020] Preferably, the sample thyristor is a packaged thyristor.

[0021] Preferably, the other end of the sample thyristor is connected to a measuring resistor, and the other end of the measuring resistor is grounded; the data acquisition device is also used to collect a voltage signal of the measuring resistor, and the processing module obtains a current signal in the main circuit through the voltage signal of the measuring resistor, and records the occurrence time of the single-particle burnout event based on the current signal.

[0022] Preferably, it also includes a clamp, by which the sample thyristor is fixed, the clamp includes two aluminum alloy plates, a first nylon plate and two second nylon plates; one side of one of the aluminum alloy plates is fixedly connected to one side of the first nylon plate, and the second nylon plates are vertically fixed at both ends of the other side of the first nylon plate, and the sample thyristor is fixedly arranged between the two aluminum alloy plates.

[0023] Preferably, an axial flow fan is fixedly provided on the outer side of one of the second nylon plates.

[0024] Preferably, the occurrence time of the single-particle burnout event and the corresponding bias voltage signal are

[0025] After fitting, the thyristor failure rate is obtained, as shown below:

[0026]

[0027] Where λ is the failure rate, V DC is the bias voltage, parameters a and b are constants, MTBF is the mean time between failures, t i is the time when the i-th single-particle burnout event occurs, t i-1 is the time when the i-1th single-particle burnout event occurs.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] The present invention proposes a thyristor failure test device. In view of the fact that the existing devices to be tested are in a non-biased state during irradiation, which does not meet the actual requirements, a power supply is set in the main circuit in the present invention to provide DC bias voltages of different sizes and polarities to the sample thyristor to meet the actual working conditions of the sample thyristor. At the same time, in view of the problem that the number of existing devices to be tested is too high and the cost is too high, the present invention sets a protection resistor in the main circuit to limit the current of the main circuit when a single particle burnout event occurs in the sample thyristor, causing the main circuit current to rise rapidly. The sample thyristor can be reused multiple times, and the frequency of SEB events of the device can be well measured while protecting the device to be tested, which greatly reduces the number of devices to be tested and reduces the test cost. Finally, the occurrence time of the single particle burnout event under DC bias voltages of different sizes and polarities and the corresponding bias voltage signal are fitted by the processing module to obtain the thyristor failure rate, which truly reflects the thyristor irradiation failure under voltage bias under actual working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 This is a schematic diagram of the cosmic ray cascade reaction in the atmosphere;

[0032] Figure 2 Schematic diagram of neutron flux changing with height;

[0033] Figure 3 A structural diagram of a thyristor failure test device of the present invention;

[0034] Figure 4 It is a schematic diagram of the clamp structure of the present invention;

[0035] Figure 5 This is a flow chart of a single irradiation test of the present invention;

[0036] Figure 6 This is the reference waveform of the SEB event.

[0037] In the figure: 1-power supply, 2-protection resistor, 3-sample thyristor, 4-artificial neutron source, 5-measuring resistor, 6-data acquisition equipment, 7-thermocouple, 8-aluminum alloy plate, 9-first screw, 10-first nylon plate, 11-second nylon plate, 12-axial flow fan, 13-second screw. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] Related background:

[0040] (1) Atmospheric neutron irradiation environment: After the primary cosmic rays in the universe enter the Earth's atmosphere, they will undergo multiple cascade interactions with the elements in the atmosphere (mainly oxygen and nitrogen), forming secondary cosmic rays in the atmosphere, which are composed of neutrons, protons, pions, muons, electrons and photons. Among them, charged particles will be shielded by the geomagnetic field and stop moving within a short distance, while neutrons are uncharged and will continue to undergo cascade spallation reactions with the atmosphere to form atmospheric neutrons. The schematic diagram of the cascade reaction of cosmic rays in the atmosphere is shown in the figure below. Figure 1 shown.

[0041] Studies have shown that the energy range of atmospheric neutrons is widely distributed, but it is mainly concentrated near fast neutrons and medium-to-high energy neutrons. In addition, the atmospheric neutron flux varies with longitude and latitude, solar activity and altitude. As the altitude increases, the neutron flux increases rapidly, with a peak value at about 20km. The neutron flux at 30km is slightly lower than the peak neutron flux; and compared with the ground, the atmospheric neutron flux at 20km is about two orders of magnitude higher, such as Figure 2 As shown. When semiconductor devices are exposed to the atmosphere, they will inevitably be irradiated by cosmic rays. Although the flux of muons is also large, it cannot explain the problem of semiconductor damage rate. At present, it is generally believed that the irradiation effect of cosmic rays on the ground is mainly atmospheric neutrons.

[0042] (2) Single-particle effect: The irradiation effect of neutrons on semiconductor devices is mainly divided into ionization effect and displacement effect. Since neutrons themselves are not charged, the ionization effect of neutrons and semiconductors is mainly achieved through the atomic ionization caused by the secondary charged particles generated in the collision process during the transport process, while the displacement effect is achieved through the direct action of neutrons on the target nucleus or the indirect action of secondary particles. The ionization effect can usually be divided into single-particle effect (Single Event Effects, SEE) and total dose effect (abbreviated as Total Ionizing Dose, TID). Among these three irradiation effects, the displacement effect and total dose effect are cumulative effects, while the single-particle effect is a transient effect.

[0043] Since there is no oxide layer structure inside the thyristor, the ionization effect of the neutron-irradiated thyristor is mainly single-particle effect. Single-particle effect can be divided into two categories according to the effect: soft error and hard error. Generally speaking, soft error has a short action time and is not destructive, while hard error has a long action time and is destructive. For power device thyristors, the impact of hard errors is mainly considered. It is mainly single event burnout (SEB), which is an important factor in neutron irradiation damage to thyristors. Single event burnout is caused by a single high-energy particle. This process causes local high current density in semiconductor devices. This type of hard error usually leads to catastrophic failure. Single event burnout mainly affects bipolar transistors and N-channel power MOSFETs in space, but the phenomenon has also been observed in high-voltage diodes in ground applications.

[0044] (3) Accelerated life test method: In 1967, the Rome Air Development Center (RADC) proposed an accelerated effect test method. On the basis of reasonable engineering and statistical assumptions, the statistical model related to the physical failure law is used to convert the reliability information obtained under an accelerated environment exceeding the normal stress level, and obtain a reproducible numerical estimate of the reliability characteristics of the test piece under the rated stress level. Accelerated life testing is a rapid life assessment test technology that shortens the test cycle by increasing the test stress level while keeping the failure mechanism unchanged. In conventional life testing, the long test cycle is the main obstacle to engineering application, and accelerated life testing provides a technical solution. Accelerated life testing technology meets the engineering needs of high-reliability and long-life evaluation, and has therefore received widespread attention in the field of weapon equipment reliability engineering.

[0045] (4) Artificial neutron source facilities: The artificial neutron irradiation source has an injection rate several orders of magnitude higher than that of atmospheric neutrons, and is an ideal simulation source for studying single-particle effects of semiconductor devices. Neutron sources usually refer to devices that produce neutrons, which generally include three categories: radioisotope neutron sources, reactor neutron sources, and accelerator neutron sources. They produce neutrons in different ways.

[0046] Radioactive isotope neutron sources use isotopes that can spontaneously fission or rays produced by isotope decay to react with other target nuclei to produce neutrons. They are characterized by small size and ease of use under various working conditions. However, the yield of isotope neutron sources is lower than that of other neutron sources, and there is a problem with their service life.

[0047] In the reactor neutron source, the use of 235 U. 239 Heavy nuclei such as Pu are made into nuclear fuel and placed in the reactor core. Neutrons are used to induce fission reactions, producing neutrons, which in turn induce fission reactions in other heavy nuclei. This fission reaction continues, forming a chain fission reaction, continuously producing neutrons and huge amounts of energy. Therefore, the advantage of a reactor neutron source is its high neutron yield, but its neutron energy spectrum is complex, there are radiation safety issues, and it is bulky and requires high construction investment, which is not conducive to widespread application.

[0048] The accelerator neutron source uses an accelerator to produce neutrons. The accelerator can accelerate charged particles in an electric field, thereby increasing their energy. The accelerated charged particles bombard the target atomic nuclei, and when the reaction conditions are met, nuclear reactions will occur to produce neutrons. The advantages of this type of neutron source are high neutron yield, low gamma background, and good radiation safety performance. It is an advanced neutron source that has been widely studied and developed in recent years.

[0049] The artificial neutron irradiation source has a fluence rate several orders of magnitude higher than that of atmospheric neutrons, making it an ideal simulation source for studying single-particle effects in semiconductor devices. Using an artificial neutron source for irradiation life testing can greatly reduce the time and cost required for life testing.

[0050] Artificial neutron accelerated irradiation experiment is an irradiation failure characteristic test of power semiconductor devices based on spallation neutron sources or 14MeV neutron generators. For the threat of atmospheric neutron irradiation of thyristors, acceleration effect tests can be carried out based on artificial neutron sources. The principle is that after establishing equivalence between the energy spectrum of artificial neutron sources and the energy spectrum of atmospheric neutrons, it can be considered that thyristors irradiated by artificial neutron sources with a certain irradiation dose within a certain period of time can be equivalent to being irradiated with a corresponding dose in an atmospheric neutron irradiation environment.

[0051] Reference Figure 3The present invention provides a thyristor failure test device for SEB damaged thyristors, comprising a power supply 1, a protective resistor 2, an artificial neutron source 4, a measuring resistor 5, a data acquisition device 6 and a thermocouple 7. The power supply 1, the protective resistor 2, the sample thyristor 3 and the measuring resistor 5 form a main circuit.

[0052] The power supply 1 is an adjustable high-voltage DC power supply, whose output terminal outputs high voltage and whose ground terminal is grounded, and is used to maintain the voltage bias stress on the sample thyristor 3 during the irradiation experiment to study the characteristics of failure rate changing with the polarity and size of the bias voltage.

[0053] One end of the protection resistor 2 is connected to the output end of the power supply 1, and the other end is connected to one end of the sample thyristor 3. When the sample thyristor 3 is irradiated and a SEB event occurs, the sample thyristor 3 is abnormally turned on, and the equivalent resistance drops sharply. At this time, the protection resistor 2 limits the maximum current in the main circuit, thereby limiting the destructive energy of the SEB event inside the sample thyristor 3, thereby achieving the purpose of protecting the sample.

[0054] The sample thyristor 3 is a device under test (DUT), and the other end of the sample thyristor 3 is connected to one end of the measuring resistor 5. The sample thyristor 3 is fixed by a fixture and accurately placed on the beam channel of the artificial neutron source.

[0055] The other end of the measuring resistor 5 is grounded. The voltage across the measuring resistor is collected by the data acquisition device 6, and the current in the main circuit, that is, the current flowing through the sample thyristor 3, can be calculated. The data acquisition device 6 can be a recorder or an oscilloscope, etc., which are not shown in the figure.

[0056] The data acquisition device 6 above has two measurement lines, one is to collect the potential difference between the high-voltage side of the sample thyristor and the ground. Since the measuring resistor 5 is much smaller than the equivalent resistance of the protection resistor 2 and the sample thyristor 3, it can be considered that the bias voltage on the sample thyristor 3 is measured. The other line measures the potential difference between the high-voltage side of the measuring resistor 5 and the ground. The measuring resistor 5 is a high-precision resistor, so the loop current can be calculated by I=U / R.

[0057] The data acquisition device also includes a thermocouple 7 for monitoring the real-time temperature data on the sample thyristor 3. Specifically, the measurement can be performed by a patch-type thermocouple. Since the low-voltage side of the sample thyristor 3 is not directly grounded, there is a weak voltage signal on the sample thyristor, and a patch with good thermal conductivity but conductive insulation is required to electrically isolate the thermocouple from the sample thyristor.

[0058] Since temperature stress will also affect the failure rate of sample thyristors damaged by radiation, the need for temperature control should be considered when designing the experimental device.

[0059] The data acquisition device below is connected to a thermocouple, which is fixed to the sample thyristor 3 by electrical insulating glue. The temperature data on the sample thyristor 3 can be collected in real time, which is a method to prove that the temperature stress is constant.

[0060] In this embodiment, the sample thyristor 3 uses a packaged thyristor instead of a thyristor chip. The specific heat of the thyristor chip is low, and its temperature will rise during the continuous repetitive experiment, and it is impossible to control the device temperature to be consistent during the entire irradiation period. The temperature has a relatively large impact on the failure rate of the thyristor under neutron irradiation conditions, so it is necessary to use a finished thyristor with a metal package for the experiment. At this time, the heat dissipation conditions are better and the temperature of the sample can be controlled more stably. Therefore, during the irradiation period, only one experimental device to be tested is allowed to be tested at the same time on the neutron path.

[0061] In this embodiment, the measuring resistor 5 should be a non-inductive resistor, and its resistance should be much smaller than the equivalent resistance of the sample thyristor 3 .

[0062] In the field circuit setting of the main circuit, the connecting wires between different devices should be as short as possible to reduce the impact of the circuit impedance on the experiment. In addition, except for the sample thyristor 3 fixed on the neutron beam channel by a clamp in the main circuit, the remaining components and wires should be placed away from the neutron irradiation path, especially in front of the sample thyristor 3, no objects should be placed to avoid affecting the neutron energy spectrum irradiated on the sample thyristor 3. At the same time, no voltage-stabilizing capacitor is set in the main circuit, because the voltage-stabilizing capacitor discharges during the internal conduction process of the device during SEB, which will cause the electrical energy to be converted into heat energy inside the device, resulting in local ablation and damage of the thyristor, which is not conducive to conducting non-destructive repeated experiments.

[0063] The present invention designs a fixture for the sample thyristor 3 to ensure stable mechanical fixation, good insulation and heat dissipation. The fixture structure is as follows: Figure 4 It includes an aluminum alloy plate 8, a first screw 9, a first nylon plate 10, two second nylon plates 11, an axial flow fan 12 and a second screw 13.

[0064] The aluminum alloy plate 8 and the first screw 9 provide a crimping force of 2 to 3 kN for the sample thyristor 3 and have a relatively fast radiation decay capability. The first nylon plate 10 and the two second nylon plates 11 play the role of insulation, auxiliary fixing and support, and an axial flow fan 12 is installed on one of the second nylon plates 11 to provide active heat dissipation to ensure that the temperature condition of the sample thyristor 3 is constant during the experiment.

[0065] The data acquisition device 6 includes a voltage probe and an oscilloscope (wave recorder or data acquisition card, etc.). The voltage probe measures the voltage signals of the sample thyristor 3 and the measuring resistor 5. During the experiment, the neutrons generated by the artificial neutron source 4 continuously bombard the sample thyristor 3, during which a single particle burnout effect (SEB) occurs inside the sample thyristor, an abnormal conductive channel is formed inside the device, and the loop current rises rapidly. At this time, a voltage pulse with a fast falling edge will be generated on the sample thyristor 3, and at the same time, the protective resistor 2 limits the loop current so that the device can be prevented from irreversible burning. After the single particle burnout event occurs, the reversible damage generated inside the device gradually recovers, and the bias voltage on the sample thyristor 3 gradually recovers to the bias state, and continues to wait for the next SEB event to occur.

[0066] In terms of SEB event signal acquisition, the falling edge voltage signal on the sample thyristor 3 and the rising edge current signal in the loop can be set as the basis for triggering the recording of waveform data, such as Figure 6 shown.

[0067] In this embodiment, some data acquisition devices 6 cannot be repeatedly triggered to record data during the process of recording waveform data and writing it into the memory after being triggered. When setting the data acquisition device, it is necessary to consider the shortest recording time between two pulses. In addition, if the data acquisition device needs to be placed in the irradiation room, it is necessary to use protective materials such as boron-containing polyethylene to protect the device to prevent the device from being damaged by irradiation interference during the neutron irradiation period.

[0068] Initially, several groups of pre-experiments should be set up to obtain the appropriate voltage limit values ​​for thyristors of specific models and specific bias directions (forward bias or reverse bias) in irradiation experiments. The voltage values ​​of the pre-experiments are set from low to high below the rated voltage of the device under test, and each group of experiments is terminated after accumulating the appropriate injection amount. The voltage limit value will be used as the upper limit reference value of the sample bias in the formal test. The judgment condition is that the irradiation experiment is carried out at this voltage, the failure data is collected as quickly as possible, the sample failure frequency is within the sampling accuracy range of the measurement system, and the initial value of the leakage current of the sample in the blocking state is always in a relatively stable healthy state during the experiment.

[0069] Since packaged thyristors are used, the oxygen-free copper, molybdenum sheet and other materials of the thyristor shell will affect the neutron energy spectrum, so only one sample is irradiated in a single experiment. After the experiment begins, the particle beam generated by the artificial neutron source continuously bombards the sample thyristor to be tested, causing random SEB events. Below the voltage limit, the experimental voltage is lowered in a certain step size to conduct multiple groups of irradiation experiments, and each group of experiments is terminated after reaching the preset irradiation dose or obtaining a sufficient number of SEB event data.

[0070] In general, the termination condition of the experimental sequence of each sample thyristor is that no valid SEB events are collected in the last experimental group, that is, the experimental voltage of this group can be preliminarily identified as the safe threshold voltage of the device under irradiation conditions. The value of the preset irradiation dose in the experiment should ensure that the final obtained device safety threshold voltage meets the reliability standard of the actual safe working life of the device. Since the failure frequency of devices based on the SEB effect increases exponentially with the experimental voltage, when the number of SEB events is used as the experimental termination condition, it should be ensured that the recorded valid number is applicable to the failure rate assessment method, otherwise the irradiation dose should be used as the termination condition.

[0071] Reference Figure 5 The principle of multiplexing samples in a single experiment to complete continuous data collection is as follows:

[0072] In order to improve the data acquisition rate and sample thyristor utilization efficiency in the experiment, the thyristor of the sample to be tested was protected by current limiting by connecting a series protective resistor to divide the voltage. Every time a SEB event occurs, since a conductive channel is about to be formed inside the sample, the voltage on the sample will be quickly transferred to both ends of the protective resistor, thereby suppressing the channel current and causing the conductive channel to be naturally shut down. This protection method will not affect the probability of SEB events, and the sample itself will not be damaged. This circuit structure allows the samples in each group of experiments to support data acquisition of dozens or hundreds of repetitive SEB events.

[0073] In this embodiment, for thyristors with different rated voltages and current carrying capacities, or for different bias states of the same thyristor, the value of the protection resistor for the irradiation experiment may be different, and should be selected in conjunction with circuit parameter research.

[0074] When the leakage current of the sample is always kept within a reasonable range during the irradiation experiment, the voltage across the sample thyristor may slightly change near the preset voltage value. This is the result of the small current fluctuation in the random event test being amplified by the protection circuit, which is the difference between the accelerated irradiation experiment of multiplexed samples and the acquisition of data in ordinary experiments.

[0075] In terms of failure rate model fitting, the data utilization efficiency of the constant stress fitting method is very low for many data points with similar but different conditions. Therefore, the present invention adopts a data processing method of variable stress fitting, that is, each SEB event collected above the safety threshold voltage in the irradiation experiment is regarded as a valid data point, the most recent stable voltage value before its occurrence is regarded as the corresponding experimental voltage, and the time interval between each two valid events is regarded as the mean time between failures corresponding to the group of data. The basis for determining a single valid SEB event is selected as a voltage drop of more than 50% across the sample.

[0076] For the experimental group that collected n valid events, assume that the data set has (n-1) failure time data timestamps Δt i (i=1,2,…,n-1), corresponding to events SEB2,SEB3,…,SEB n Obviously, each failure time data is the time interval between two valid events, which is the corresponding mean time between failures MTBF. (i-1) and SEB i , event SEB (i-1) After the occurrence of the thyristor voltage is V (i-1)A , event SEB i The voltage on the thyristor before the event occurs is V iB , then the bias voltage V i =f(V (i-1)A ,V iB The mean time between failures is MTBF(V i )=Δt i =t i -t (i-1) . After calculation and verification, when the bias voltage V i The value is the voltage stability value before the event V iB The optimal combination of computational efficiency and accuracy can be obtained. Combining the exponential failure model proposed by foreign researchers, the estimated formula for the effect of bias voltage on device failure rate can be obtained as follows:

[0077]

[0078] Where λ is the failure rate, V DC is the DC bias voltage, and parameters a and b are constants. When the temperature is considered constant, based on the experimentally obtained (λ, V DC ) data pair, the parameters a and b can be fitted and solved, thus obtaining the relationship between the SEB failure rate of the device under test irradiated by neutrons and the change of bias voltage. Obviously, the experimental data pair (Δt i ,V i ) the more it is, the better the fitting result of this method will be.

[0079] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0080] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A thyristor failure test device, characterized in that: include: A power supply (1) is used to provide a DC bias voltage of different magnitudes and polarities to a sample thyristor (3); An artificial neutron source (4) is used to generate neutron beams with different irradiation dose rates, and irradiate the sample thyristor (3) with the neutron beams, so that a random single particle burning event occurs inside the sample thyristor (3); A protective resistor (2), one end of which is connected to the output end of the power supply (1), and the other end of which is connected to one end of the sample thyristor (3). When a single particle burnout event occurs in the sample thyristor (3), a conductive channel is formed inside the sample thyristor (3), causing the main circuit current to rise. At this time, the protective resistor (2) is used to limit the main circuit current. A data acquisition device (6), electrically connected to the sample thyristor (3), and used for real-time acquisition of a bias voltage signal on the sample thyristor (3); The processing module is used for recording the occurrence time of the single-particle burnout event according to the bias voltage signal on the sample thyristor (3) at this time when a single-particle burnout event occurs in the sample thyristor (3) under direct current bias voltages of different magnitudes and polarities, and performing fitting processing on multiple occurrence times of the single-particle burnout event and the corresponding bias voltage signals to obtain the thyristor failure rate.

2. A thyristor failure test device as claimed in claim 1, characterized in that: The sample thyristor (3) is arranged in a neutron beam channel of an artificial neutron source (4).

3. A thyristor failure test device as claimed in claim 1, characterized in that: The data acquisition device (6) comprises a voltage probe and an oscilloscope.

4. A thyristor failure test device as claimed in claim 1, characterized in that: A thermocouple (7) is provided on the outside of the sample thyristor (3), and the thermocouple (7) is used to measure the temperature of the sample thyristor (3).

5. A thyristor failure test device as claimed in claim 1, characterized in that: The sample thyristor (3) is a packaged thyristor.

6. A thyristor failure test device as claimed in claim 1, characterized in that: The other end of the sample thyristor (3) is connected to a measuring resistor (5), and the other end of the measuring resistor (5) is grounded; the data acquisition device (6) is also used to collect a voltage signal of the measuring resistor (5); the processing module obtains a current signal in the main circuit through the voltage signal of the measuring resistor (5), and records the occurrence time of the single particle burnout event based on the current signal.

7. A thyristor failure test device as claimed in claim 1, characterized in that: The invention also comprises a clamp, by which the sample thyristor (3) is fixed, the clamp comprising two aluminum alloy plates (8), a first nylon plate (10) and two second nylon plates (11); one side of one of the aluminum alloy plates (8) is fixedly connected to one side of the first nylon plate (10), and the second nylon plates (11) are vertically fixed at both ends of the other side of the first nylon plate (10), and the sample thyristor (3) is fixedly arranged between the two aluminum alloy plates (8).

8. A thyristor failure test device as claimed in claim 7, characterized in that: An axial flow fan (12) is fixedly provided on the outer side of one of the second nylon plates (11).

9. A thyristor failure test device as claimed in claim 1, characterized in that: The occurrence time of the single-particle burnout event and the corresponding bias voltage signal are fitted to obtain the thyristor failure rate, which is specifically as follows: Where λ is the failure rate, V DC is the bias voltage, parameters a and b are constants, MTBF is the mean time between failures, t i is the time when the i-th single-particle burnout event occurs, t i-1 is the time when the i-1th single-particle burnout event occurs.

Images