CN116840670A - Switch life assessment method of GaN power conversion device - Google Patents
Switch life assessment method of GaN power conversion device Download PDFInfo
- Publication number
- CN116840670A CN116840670A CN202310638623.0A CN202310638623A CN116840670A CN 116840670 A CN116840670 A CN 116840670A CN 202310638623 A CN202310638623 A CN 202310638623A CN 116840670 A CN116840670 A CN 116840670A
- Authority
- CN
- China
- Prior art keywords
- life
- test
- gan power
- stress
- acceleration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000006243 chemical reaction Methods 0.000 title claims abstract description 58
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000012360 testing method Methods 0.000 claims abstract description 139
- 230000001133 acceleration Effects 0.000 claims abstract description 108
- 238000007405 data analysis Methods 0.000 claims abstract description 4
- 239000003990 capacitor Substances 0.000 claims description 27
- 239000008186 active pharmaceutical agent Substances 0.000 claims description 24
- 238000010586 diagram Methods 0.000 claims description 18
- 230000004913 activation Effects 0.000 claims description 10
- 238000010998 test method Methods 0.000 claims description 9
- 238000005315 distribution function Methods 0.000 claims description 8
- 230000001939 inductive effect Effects 0.000 claims description 6
- 238000007476 Maximum Likelihood Methods 0.000 claims description 4
- JMASRVWKEDWRBT-UHFFFAOYSA-N Gallium nitride Chemical compound [Ga]#N JMASRVWKEDWRBT-UHFFFAOYSA-N 0.000 description 96
- 229910002601 GaN Inorganic materials 0.000 description 95
- 230000007246 mechanism Effects 0.000 description 10
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 9
- 229910052710 silicon Inorganic materials 0.000 description 9
- 239000010703 silicon Substances 0.000 description 9
- 230000015556 catabolic process Effects 0.000 description 4
- 239000004065 semiconductor Substances 0.000 description 4
- 238000006731 degradation reaction Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000014509 gene expression Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 239000012084 conversion product Substances 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005669 field effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 230000000638 stimulation Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/327—Testing of circuit interrupters, switches or circuit-breakers
Abstract
The invention provides a switch life assessment method of a GaN power conversion device, which comprises the following steps: adopting an independent stress source to perform a stress-failure test on a GaN power switch tube in a GaN power conversion device sample, and performing data analysis on test results to obtain a current stress acceleration life model, a voltage stress acceleration life model and a temperature stress acceleration life model; combining the current stress acceleration life model, the voltage stress acceleration life model and the temperature stress acceleration life model to obtain a multi-factor acceleration life model; performing an acceleration test on the tested GaN power conversion device to obtain an acceleration test life T Acceleration test Substituting an actual stress source of the task section into a multi-factor acceleration life model to obtain an acceleration factor AF, and multiplying the acceleration factor by an acceleration test life T Acceleration test The minimum service life of the mission profile is predicted. The invention can evaluate the switching life of the GaN power conversion device more accurately and pointedly.
Description
Technical Field
The invention relates to a method for evaluating the switch life of a GaN power converter, which is suitable for evaluating the switch life of a planar enhancement mode GaN power converter and a depletion mode GaN power converter.
Background
Gallium nitride (GaN) power switches are important devices for efficient, small power conversion applications. Current mode GaN power switches are based on planar High Electron Mobility Transistors (HEMTs), a type of Field Effect Transistor (FET). GaN power transistors have small input and output capacitances, their switching is typically very fast, on the order of nanoseconds, and although the thermal effects induced by such fast switching are small, due to the high V DS (GaN Power switch drain to Source Voltage) and I DP (drain current in the on state of the GaN power switch) induces higher electrical stress. Therefore, for GaN power transistors, characterization of switch reliability is also necessary in addition to conventional silicon power transistor safe operating area SOA testing.
Aiming at the actual demands of aerospace engineering such as deep space exploration, broadband communication satellites and the like on an aerospace power management chip with high ionization radiation resistance, light weight, high efficiency and high reliability and long service life, a GaN wide forbidden band semiconductor has the advantages of high breakdown electric field, high thermal conductivity, high electron saturation rate and high radiation resistance, and becomes the most preferable of the technical scheme of the next generation of aerospace power devices. However, the existing silicon power transistor Safe Operating Area (SOA) test method is not enough to fully characterize a GaN power transistor, the switching speed of a GaN device is very fast, and voltage and current stresses are experienced simultaneously in a plurality of switching processes; in addition to the conventional SOA test, gaN power transistors need to be tested under continuous switching conditions. Silicon-based power conversion product testing is either inapplicable or not yet fully specified for GaN switches, only in relevant literature, such as document 1s.singhal et al, "quality qualification and reliability of one GaN process platform"; document 2e.zanoni et al, "AIGAN/GaN based HEMT failure physics and reliability: the mechanism affecting the gate edge and schottky junction "does not include switch life assessment.
The reliability evaluation of the GaN power conversion device by using the traditional SOA test method has larger error and has the following defects:
1) The silicon device has no failure mechanism related to such dynamic switching behavior, and the second key new degradation mechanism of GaN devices is to switch SOA (also called dynamic high temperature operating life or DHTOL), and conventional silicon-based verification test methods cannot cover GaN technology identification requirements.
2) The stress application modes are various, the switch track has correlation with the failure mode, and no widely-covered test method and standardized test operation exist at present.
3) In order to obtain a stress program, a life model of a common failure mechanism of a device needs to be mastered, comparison calculation is carried out through a plurality of task section applicable models, the most conservative model is adopted to verify the reliability of GaN running under application conditions, but GaN life cycle modeling is a developed field, and an implementation method of empirical model development and data collection and presentation is still random Fan Lujing.
Therefore, the existing MOSFET device qualification evaluation method cannot accurately evaluate the GaN MOSFET device switching life.
Disclosure of Invention
The technical problems solved by the invention are as follows: in order to overcome the practical problem that a silicon power transistor Safety Operating Area (SOA) test method is insufficient for comprehensively representing a GaN power transistor, the invention provides a switch life assessment method of a GaN power conversion device, and the method can be used for accurately and pointedly assessing the switch life of the GaN power conversion device.
The technical scheme of the invention is as follows: a method of switching life assessment for a GaN power conversion device, the method comprising the steps of:
adopting an independent stress source to perform a stress-failure test on a GaN power switch tube in a GaN power conversion device sample, and performing data analysis on test results to obtain a current stress acceleration life model, a voltage stress acceleration life model and a temperature stress acceleration life model;
combining the current stress acceleration life model, the voltage stress acceleration life model and the temperature stress acceleration life model to obtain a multi-factor acceleration life model;
performing an acceleration test on the tested GaN power conversion device to obtain an acceleration test life T Acceleration test Substituting an actual stress source of the task section into a multi-factor acceleration life model to obtain an acceleration factor AF, and multiplying the acceleration factor by an acceleration test life T Acceleration test The minimum service life of the mission profile is predicted.
Preferably, the current stress acceleration life model, the voltage stress acceleration life model and the temperature stress acceleration life model are obtained by the following methods:
preparing N GaN power conversion devices as test samples; the switching application types of the GaN power switch tubes in the N GaN power conversion device samples cover three switching types: the hard switch, the soft switch and the resistance switch are arranged, and N is more than or equal to 45;
aiming at the GaN power switch tube in each GaN power conversion device sample, selecting a corresponding continuous switch test circuit to respectively develop a temperature stress-failure test, a voltage stress-failure test and a current stress-failure test, and measuring failure characteristic parameters of the GaN power switch tube in real time; the stress-failure test of each independent stress source at least comprises three different test conditions, under each test condition, the independent stress source selects different values according to a preset gradient, and the rest stress sources are unchanged;
according to the failure characteristic parameters of the GaN power switch tube measured under each test condition, the failure time under each test condition is counted;
estimating a life distribution function under 95% confidence by adopting a maximum likelihood method according to the failure time under each test condition, fitting to obtain an optimal life distribution function under each test condition, drawing a failure time life probability distribution diagram, obtaining a life B10 with 90% reliability according to the failure time life probability distribution diagram, and drawing a relation diagram of the life B10 with 90% reliability and a stress source;
according to a relation diagram of the life B10 with 90% reliability and a stress source, obtaining effective activation energy of the switching loss, substituting the effective activation energy of the switching loss into a current stress acceleration life model, a voltage stress acceleration life model and a temperature stress acceleration life model to correct the current stress acceleration life model, the voltage stress acceleration life model and the temperature stress acceleration life model.
Preferably, the temperature stress acceleration lifetime model is an arrhenius lifetime model.
Preferably, when the GaN power conversion device sample is a hard switch, the continuous switch test circuit is an inductive load switch circuit, and the inductive load switch circuit comprises a capacitor C1-1, a diode D1-1, a resistor R1-1 and an inductor L1-1;
the capacitor C1-1 is connected between the positive electrode of the power supply and the ground in a bridging manner, the negative electrode of the diode D1-1 is connected with the positive electrode of the power supply, one end of the resistor R1-1 is connected with the positive electrode of the power supply, the other end of the resistor R1-1 is connected with the inductor L1-1 in series, the other end of the inductor L1-1 and the positive electrode of the diode D1-1 are connected with the drain electrode of the GaN power switch tube, and the grid electrode and the source electrode of the GaN power switch tube are grounded.
Preferably, when the GaN power conversion device sample is a soft switch, the continuous switch test circuit is a half-bridge configured buck conversion circuit, and the half-bridge configured buck converter comprises a capacitor C2-1, a capacitor C2-2, a CMOS tube, a resistor R2-1 and an inductor L2-1;
the capacitor C2-1 is connected between the power supply anode and the ground in a bridging manner, the drain electrode of the CMOS tube is connected with the power supply anode, the source electrode of the CMOS tube is connected with the source electrode of the GaN power switch tube, the grid electrode of the GaN power switch tube and the drain electrode of the GaN power switch tube are grounded, the grid electrode of the CMOS tube is connected with one end of the inductor L2-1, the other end of the inductor L2-1 is connected with the capacitor C2-2 and the resistor R2-1 in parallel, and the other ends of the capacitor C2-2 and the resistor R2-1 are grounded.
Preferably, for when the GaN power conversion device sample is a resistance switch, the continuous switch test circuit is a resistive load switch circuit, and the resistive load switch circuit comprises a capacitor C3-1 and a resistor R3-1;
the capacitor C3-1 is connected between the power supply anode and the ground in a bridging way, one end of the resistor R3-1 is connected with the power supply anode, the other end is connected with the drain electrode of the GaN power switch tube, the grid electrode of the switch tube is grounded, the source electrode of the GaN power switch tube is connected with the resistor R3-1, and the drain electrode is grounded.
Preferably, the failure characteristic parameter comprises measurement V DS 、R DS(ON) 、I DP ；
And when any failure characteristic parameter drifts out of a range determined by a preset minimum value and a preset maximum value, defining the failure characteristic parameter as failure, and recording the failure time.
Preferably, the test conditions for the current stress-failure test are set as follows:
controlling the voltage between the drain electrode and the source electrode of the GaN power switch tube to keep the rated voltage value unchanged, and keeping the junction temperature to a preset value according to a preset gradient value;
the test conditions for the voltage stress-failure test were set as follows:
controlling the current between the drain electrode and the source electrode of the GaN power switch tube to keep the rated current unchanged, and keeping the junction temperature at a preset value according to a preset gradient value of the voltage;
the test conditions for the temperature stress-failure test were set as follows:
and keeping the rated current of the current between the drain electrode and the source electrode in the on state of the GaN power switch tube or keeping the rated voltage of the voltage between the drain electrode and the source electrode in the off state of the GaN power switch tube unchanged, and taking the value of the junction temperature according to a preset gradient.
Preferably, the multi-factor accelerated life model is:
AF＝K 1 ·f v (V DS )f i (I DP )f T (T)
wherein AF is an acceleration factor; f (f) v (V DS ) F is a voltage stress acceleration life model i (I DP ) F is a current stress acceleration life model T (T) is a temperature stress acceleration life model, V DS For the drain-to-source voltage of GaN power switch tube, I DP The drain current in the on state of the GaN power switch is T, the junction temperature of the GaN power switch tube is K 1 And accelerating the life model coefficient for multiple factors.
The current stress-failure test, the temperature stress-failure test and the voltage stress-failure test are all carried out by adopting a test method of a fixed number of truncations until all circuits fail, and the test is stopped.
Compared with the prior art, the invention has the beneficial effects that:
(1) The present invention classifies by switching pressure stimuli and their trajectories and uses more severe stress conditions to cover low application conditions, resulting in a widely covered program.
(2) The present invention provides a procedure for obtaining a wear model and device lifetime.
(3) The present invention provides general guidelines for selecting test circuits and developing test plans.
Drawings
FIG. 1 is a flow chart of a method for evaluating the switching life of a GaN power conversion device of the invention;
FIG. 2 (a) shows a tunable I of a hard switch according to an embodiment of the invention DP -V DS A track;
FIG. 2 (b) is a tunable I of a soft switch according to an embodiment of the invention DP -V DS A track;
FIG. 2 (c) is a diagram of a tunable I of a resistive switch according to an embodiment of the present invention DP -V DS A track;
FIG. 3 (a) shows an inductive load switching circuit according to an embodiment of the present invention;
FIG. 3 (b) is a synchronous buck converter circuit according to an embodiment of the present invention;
FIG. 3 (c) shows a resistive load switching circuit according to an embodiment of the present invention;
FIG. 4 is a drawing showing a test event record in accordance with an embodiment of the present invention;
FIG. 5 is a graph showing a lifetime failure probability distribution in accordance with an embodiment of the present invention.
Detailed Description
The following describes in further detail the embodiments of the present invention with reference to the drawings and examples. The following examples are illustrative of the invention and are not intended to limit the scope of the invention.
Fig. 1 is a flowchart of a switching life evaluation method of a GaN power conversion device of the invention. As shown in fig. 1, the method for evaluating the switching life of the GaN power conversion device according to the embodiment of the invention includes the steps of:
1. adopting an independent stress source to perform a stress-failure test on a GaN power switch tube in a GaN power conversion device sample, and performing data analysis on test results to obtain a current stress acceleration life model, a voltage stress acceleration life model and a temperature stress acceleration life model;
1.1, 45 GaN power conversion devices are prepared as test samples, the numbers of each GaN power conversion device are recorded, and the GaN power conversion devices are randomly divided into 9 groups.
The switching types of the GaN power switching tubes in the 45 GaN power conversion device samples cover three switching types: hard switches, soft switches, and resistive switches;
in this embodiment, the switch type of the circuit to be evaluated is first determined, and the correlation between the switch trace and the failure mode is considered, I DP -V DS The trace profile of the waveform during the switching cycle can classify switching stimuli that cause different levels or types of device degradation.
The switch type is determined according to the following method:
s1.1.1, obtaining I obtained by a GaN power conversion device under pressure stimulation of a preset switching period DP -V DS A trajectory curve;
s1.1.2 when the voltage and the current are not 0, the voltage and the current overlap, and the voltage and the current waveforms are overcharged, so that switching noise is generated, the GaN power conversion device belongs to a hard switch; as shown in fig. 2 (a);
when the voltage drop at the two ends of the tested device is 0 before the tested device is turned on or the current flowing through the tested device is 0 before the tested device is turned off, the GaN power conversion device belongs to a soft switch; as shown in fig. 2 (b);
when no current peak exists in the switching waveform, the GaN power conversion device belongs to the waveform and track of the resistive load, and the GaN power conversion device belongs to the resistance switch. As shown in fig. 2 (c);
1.2, aiming at the GaN power switch tube in each GaN power conversion device sample, selecting a corresponding continuous switch test circuit to respectively carry out a temperature stress-failure test, a voltage stress-failure test and a current stress-failure test, and measuring failure characteristic parameters of the GaN power switch tube in real time; the stress-failure test of each independent stress source at least comprises three different test conditions, under each test condition, the independent stress source selects different values according to a preset gradient, and the rest stress sources are unchanged;
based on a switching stress application profile of a switching type in device application, selecting corresponding continuous switching test circuits for respectively developing a temperature stress-failure test, a voltage stress-failure test and a current stress-failure test aiming at GaN power switching tubes in samples of each GaN power conversion device, and measuring failure characteristic parameters of the GaN power switching tubes in real time.
The hard switching stress type may use the inductive load switching circuit of fig. 3 (a), which includes a capacitor C1-1, a diode D1-1, a resistor R1-1, an inductor L1-1;
the capacitor C1-1 is connected between the positive electrode of the power supply and the ground in a bridging manner, the negative electrode of the diode D1-1 is connected with the positive electrode of the power supply, one end of the resistor R1-1 is connected with the positive electrode of the power supply, the other end of the resistor R1-1 is connected with the inductor L1-1 in series, the other end of the inductor L1-1 and the positive electrode of the diode D1-1 are connected with the drain electrode of the GaN power switch tube, and the grid electrode and the source electrode of the GaN power switch tube are grounded.
The soft switching stress type can be selected from the half-bridge configured buck converter of fig. 3 (b), wherein the half-bridge configured buck converter comprises a capacitor C2-1, a capacitor C2-2, a CMOS tube, a resistor R2-1 and an inductor L2-1;
the capacitor C2-1 is connected between the power supply anode and the ground in a bridging manner, the drain electrode of the CMOS tube is connected with the power supply anode, the source electrode of the CMOS tube is connected with the source electrode of the GaN power switch tube, the grid electrode of the GaN power switch tube and the drain electrode of the GaN power switch tube are grounded, the grid electrode of the CMOS tube is connected with one end of the inductor L2-1, the other end of the inductor L2-1 is connected with the capacitor C2-2 and the resistor R2-1 in parallel, and the other ends of the capacitor C2-2 and the resistor R2-1 are grounded.
When the GaN power conversion device sample is a resistance switch, the continuous switch test circuit is a resistive load switch circuit, and the resistive load switch circuit can be shown in the figure 3 (C), and comprises a capacitor C3-1 and a resistor R3-1;
the capacitor C3-1 is connected between the power supply anode and the ground in a bridging way, one end of the resistor R3-1 is connected with the power supply anode, the other end is connected with the drain electrode of the GaN power switch tube, the grid electrode of the switch tube is grounded, the source electrode of the GaN power switch tube is connected with the resistor R3-1, and the drain electrode is grounded.
Based on the switching stress application profile of the switching position type in the device application, a test platform is built, and the built test platform allows for accelerated testing under the condition of being higher than the design voltage and current and has the capabilities of measuring and controlling the temperature of the device and acquiring waveforms and measuring parameters.
The number of GaN power conversion devices N is a multiple of 9, and the more preferably the same batch of Devices Under Test (DUTs), the more limited such devices are damaged for failure analysis of the damaged devices in the test.
The test conditions for the current stress-failure test were set as follows:
controlling the voltage between the drain electrode and the source electrode of the GaN power switch tube to keep the rated voltage value unchanged, and keeping the junction temperature to a preset value according to a preset gradient value; at least the current
The test conditions for the voltage stress-failure test were set as follows:
controlling the current between the drain electrode and the source electrode of the GaN power switch tube to keep the rated current unchanged, and keeping the junction temperature at a preset value according to a preset gradient value of the voltage;
the test conditions for the temperature stress-failure test were set as follows:
and keeping the rated current of the current between the drain electrode and the source electrode in the on state of the GaN power switch tube or keeping the rated voltage of the voltage between the drain electrode and the source electrode in the off state of the GaN power switch tube unchanged, and taking the value of the junction temperature according to a preset gradient.
The present invention focuses on each source independently by keeping other factors unchanged, such as: temperature acceleration, voltage, current, etc., each stress condition is at least 5 samples in size and at least 3 stress conditions for device failure due to a single wear mechanism.
In this embodiment, 9 groups of circuits are respectively allocated to a current stress-failure test, a temperature stress-failure test and a voltage stress-failure test, each group of tests is provided with 3 test conditions, and the current stress-failure test, the temperature stress-failure test and the voltage stress-failure test are all carried out by adopting a test method of a fixed number of truncations until all the circuits fail, and the test is stopped.
1.3, counting the failure time under each test condition according to the failure characteristic parameters of the GaN power switch tube measured under each test condition;
at the beginning and end of the stress test, V should be measured DS 、R DS(ON) Key measured parameters such as gate leakage, drain leakage, etc. The failure characteristic parameter includes a measurement V DS 、R DS(ON) Gate leakage, drain leakage;
and when any failure characteristic parameter drifts out of a range determined by a preset minimum value and a preset maximum value, defining the failure characteristic parameter as failure, and recording the failure time.
1.4, estimating a life distribution function under 95% confidence coefficient by adopting a maximum likelihood method according to the failure time under each test condition, fitting to obtain an optimal life distribution function under each test condition, drawing a failure time life probability distribution diagram, obtaining a life B10 with 90% reliability according to the failure time life probability distribution diagram, and drawing a relation diagram of the life B10 with 90% reliability and a stress source;
according to failure data of 3-level gradient test conditions under three test types, a test event record chart is shown in figure 4, according to failure time under each test condition, a maximum likelihood method is adopted to estimate parameters of a life distribution function under 95% confidence, fitting is carried out in Weibull distribution, lognormal distribution and logarithmic distribution multi-type functions, an optimal life distribution function is selected, and a failure time life probability distribution chart is drawn, wherein the graph is shown in figure 5; according to the failure time life probability distribution diagram, obtaining a life B10 with 90% reliability, and drawing a relation diagram of the life B10 with 90% reliability and stress;
the abscissa of the failure time life probability distribution diagram is failure time, the ordinate is accumulated failure probability, and the accumulated failure probability in the diagram is the failure life corresponding to 10%, namely the life with 90% reliability.
And 1.5, obtaining effective activation energy of the switching loss according to a relation diagram of the life B10 with 90% reliability and the stress source, and substituting the effective activation energy of the switching loss into a current stress acceleration life model, a voltage stress acceleration life model and a temperature stress acceleration life model to correct the current stress acceleration life model, the voltage stress acceleration life model and the temperature stress acceleration life model.
In this embodiment, the on-state current or the off-state voltage is kept at its rated value, other factors are kept unchanged, stress-failure tests are performed at least three different temperatures, a failure time life distribution diagram of each temperature is drawn, required failure time is obtained, ln < ttf > vs 1/T is drawn, effective activation energy of switching loss is obtained, and an arrhenius life model of the device is corrected and built.
In this embodiment, the on-state current is maintained at its nominal value, the junction temperature is maintained at a suitable fixed use condition value, other factors are kept unchanged, stress-failure tests are performed at three or more different voltages, the failure time life profile of each voltage is plotted, the desired failure time is obtained, the relationship between fv-1< ttf > and VDP is plotted, and a suitable voltage stress wear model is obtained.
In this embodiment, the off-state voltage is maintained at its nominal value, the junction temperature is fixed at a suitable fixed use condition value and other factors, stress-failure tests are performed in three or more different power-on states, and each I is plotted DP And obtaining the required failure time, and drawing fi -1 <ttf>And I DP And an appropriate wear model of the current stress is obtained.
The voltage stress acceleration life model and the current stress acceleration life model are both based on failure mechanisms and are selected according to JEDEC JEP122H-2016 failure mechanism and model of semiconductor device.
Taking the accelerated temperature test as an example, the gate oxide of a silicon device to fail due to excessive temperature can be described by the Arrhenius equation. The characteristic activation energy (Ea) can be extracted from the slope of the curve of the failure life vs. temperature value.
The Arrhenius equation is commonly used to describe the failure behavior of the gate oxide of a silicon device under temperature and bias stress, and in some cases, also for GaN failure modes.
Where dM/dt is the rate of degradation of the sensitive parameter; a is a pending parameter, ea is an activation energy (eV) boltzmann constant k=8.617×10 -5 eV/. Degree.C.T is absolute temperature (K). The constant k is the Boltzmann constant we are familiar with. Taking the logarithm to obtain:
and xi is the service life of the device, the logarithm of the service life and the reciprocal of the temperature are in linear relation, and the formula of the temperature acceleration coefficient AF (T) is as follows:
T 1 is the temperature in the case of use.
Taking one of the data points as a reference (T and failure life) and selecting a different specified use temperature (T 1 ) And calculating a temperature acceleration coefficient AF (T) and the failure life, and obtaining the activation energy Ea by fitting software.
The acceleration current test and the acceleration voltage test are similar to the step, and the acceleration model can be calculated based on the failure mechanism according to JEDEC JEP122H-2016 failure mechanism and model of semiconductor device.
2. Combining the current stress acceleration life model, the voltage stress acceleration life model and the temperature stress acceleration life model to obtain a multi-factor acceleration life model;
since the wear life model is in most semiconductor reliability descriptions, each pressure is independent of the other pressures. In this case, we can obtain the acceleration lifetime model of each stress source independently, and obtain the acceleration lifetime models of all stress sources, taking the product of lifetime expressions of each stress source as a multi-factor empirical acceleration lifetime model. This can be used to predict the lifetime of the device under typical use conditions. The following is a general equation representation of the empirical expression for the multifactor acceleration factor.
The multi-factor accelerated life model is:
AF＝K 1 ·f v (V DS )·f i (I DP )·f T (T)
wherein AF is an acceleration factor; f (f) v (V DS ) F is a voltage stress acceleration life model i (I DP ) F is a current stress acceleration life model T (T) is a temperature stress acceleration life model, V DS For the drain-to-source voltage of GaN power switch tube, I DP The drain current in the on state of the GaN power switch is T, the junction temperature of the GaN power switch tube is K 1 And fitting the multi-factor accelerated life model coefficients according to a plurality of groups of tests by the same method.
3. Performing an acceleration test on the tested GaN power conversion device to obtain an acceleration test life T Accelerated test life Substituting an actual stress source of the task section into a multi-factor acceleration life model to obtain an acceleration factor AF, and multiplying the acceleration factor by an acceleration test life T Accelerated test life The minimum service life of the mission profile is predicted.
After the accelerated life models of various stress sources and related factors are obtained and the task profile is determined, it is possible to predict the minimum service life of the profile and use the minimum service life of the profile as a measure of the reliability of the GaN power switch.
Minimum service life T of task profile Service life of application The calculation formula of (2) is specifically as follows:
T service life of application ＝T Accelerated test life ×AF (5)
Compared with the traditional silicon power transistor Safe Operating Area (SOA) test method, the GaN power conversion device switch life assessment method has the following characteristics:
1) Extensive coverage testing was achieved by classifying the switching pressure stimuli and their switching trajectories and interpreting the use of more severe pressure conditions to cover milder use case conditions (three levels).
2) The invention uses the switching tube of the power conversion device of the proper type to carry out the accelerated life test so as to generate a model.
3) By using stress condition running components in the application environment, a program that runs reliably under application use conditions is verified, which if passed, will ensure reliable operation under a wide range of use conditions.
The method takes into account the advantages and limitations of circuit boards, hardware and components, while also recognizing the necessity of not unnecessarily burdening the industry. It does not limit the stress conditions for additional testing, larger sample sizes and higher accelerations. Nor does it limit the use of circuit boards and device types. For example, by verifying that the complexity does not mask the acceleration of the relevant failure mechanism, a more complex board or GaN switch may be used to determine lifetime. Instead, a simpler circuit board may be used, ensuring application reliability by providing auxiliary materials that show robustness to untested modes of operation.
If at a higher V DS Under execution reliability verification, conditions of lower pressure are covered, and the user does not need to rerun the reliability test under these conditions, and more severe pressure conditions can cover milder use case conditions, thereby obtaining a widely covered program.
Although the present invention has been described in terms of the preferred embodiments, it is not intended to be limited to the embodiments, and any person skilled in the art can make any possible variations and modifications to the technical solution of the present invention by using the methods and technical matters disclosed above without departing from the spirit and scope of the present invention, so any simple modifications, equivalent variations and modifications to the embodiments described above according to the technical matters of the present invention are within the scope of the technical matters of the present invention.
Claims (10)
1. The method for evaluating the switch life of the GaN power conversion device is characterized by comprising the following steps of:
adopting an independent stress source to perform a stress-failure test on a GaN power switch tube in a GaN power conversion device sample, and performing data analysis on test results to obtain a current stress acceleration life model, a voltage stress acceleration life model and a temperature stress acceleration life model;
combining the current stress acceleration life model, the voltage stress acceleration life model and the temperature stress acceleration life model to obtain a multi-factor acceleration life model;
performing an acceleration test on the tested GaN power conversion device to obtain an acceleration test life T Accelerated test life Substituting an actual stress source of the task section into a multi-factor acceleration life model to obtain an acceleration factor AF, and multiplying the acceleration factor by an acceleration test life T Accelerated test life The minimum service life of the mission profile is predicted.
2. The method for evaluating the switching life of the GaN power conversion device according to claim 1, wherein the current stress acceleration life model, the voltage stress acceleration life model and the temperature stress acceleration life model are obtained by the following steps:
preparing N GaN power conversion devices as test samples; the switching application types of the GaN power switch tubes in the N GaN power conversion device samples cover three switching types: the hard switch, the soft switch and the resistance switch are arranged, and N is more than or equal to 45;
aiming at the GaN power switch tube in each GaN power conversion device sample, selecting a corresponding continuous switch test circuit to respectively develop a temperature stress-failure test, a voltage stress-failure test and a current stress-failure test, and measuring failure characteristic parameters of the GaN power switch tube in real time; the stress-failure test of each independent stress source at least comprises three different test conditions, under each test condition, the independent stress source selects different values according to a preset gradient, and the rest stress sources are unchanged;
according to the failure characteristic parameters of the GaN power switch tube measured under each test condition, the failure time under each test condition is counted;
estimating a life distribution function under 95% confidence by adopting a maximum likelihood method according to the failure time under each test condition, fitting to obtain an optimal life distribution function under each test condition, drawing a failure time life probability distribution diagram, obtaining a life B10 with 90% reliability according to the failure time life probability distribution diagram, and drawing a relation diagram of the life B10 with 90% reliability and a stress source;
according to a relation diagram of the life B10 with 90% reliability and a stress source, obtaining effective activation energy of the switching loss, substituting the effective activation energy of the switching loss into a current stress acceleration life model, a voltage stress acceleration life model and a temperature stress acceleration life model to correct the current stress acceleration life model, the voltage stress acceleration life model and the temperature stress acceleration life model.
3. The method for evaluating the switching life of the GaN power conversion device according to claim 2, wherein the temperature stress acceleration life model is an arrhenius life model.
4. The method for evaluating the switching life of the GaN power conversion device according to claim 2, wherein when the GaN power conversion device sample is a hard switch, the continuous switch test circuit is an inductive load switch circuit, and the inductive load switch circuit comprises a capacitor C1-1, a diode D1-1, a resistor R1-1 and an inductor L1-1;
the capacitor C1-1 is connected between the positive electrode of the power supply and the ground in a bridging manner, the negative electrode of the diode D1-1 is connected with the positive electrode of the power supply, one end of the resistor R1-1 is connected with the positive electrode of the power supply, the other end of the resistor R1-1 is connected with the inductor L1-1 in series, the other end of the inductor L1-1 and the positive electrode of the diode D1-1 are connected with the drain electrode of the GaN power switch tube, and the grid electrode and the source electrode of the GaN power switch tube are grounded.
5. The method for evaluating the switching life of the GaN power conversion device according to claim 2, wherein when the GaN power conversion device sample is a soft switch, the continuous switch test circuit is a half-bridge configured buck converter circuit, and the half-bridge configured buck converter comprises a capacitor C2-1, a capacitor C2-2, a CMOS tube, a resistor R2-1 and an inductor L2-1;
the capacitor C2-1 is connected between the power supply anode and the ground in a bridging manner, the drain electrode of the CMOS tube is connected with the power supply anode, the source electrode of the CMOS tube is connected with the source electrode of the GaN power switch tube, the grid electrode of the GaN power switch tube and the drain electrode of the GaN power switch tube are grounded, the grid electrode of the CMOS tube is connected with one end of the inductor L2-1, the other end of the inductor L2-1 is connected with the capacitor C2-2 and the resistor R2-1 in parallel, and the other ends of the capacitor C2-2 and the resistor R2-1 are grounded.
6. The method for evaluating the switching life of the GaN power conversion device according to claim 2, wherein for the case that the GaN power conversion device sample is a resistance switch, the continuous switch test circuit is a resistive load switch circuit, and the resistive load switch circuit comprises a capacitor C3-1 and a resistor R3-1;
the capacitor C3-1 is connected between the power supply anode and the ground in a bridging way, one end of the resistor R3-1 is connected with the power supply anode, the other end is connected with the drain electrode of the GaN power switch tube, the grid electrode of the switch tube is grounded, the source electrode of the GaN power switch tube is connected with the resistor R3-1, and the drain electrode is grounded.
7. The method for evaluating the switching life of a GaN power conversion device according to claim 2, wherein said failure characteristic parameter comprises measuring V DS 、R DS(ON) 、I DP ；
And when any failure characteristic parameter drifts out of a range determined by a preset minimum value and a preset maximum value, defining the failure characteristic parameter as failure, and recording the failure time.
8. The method for evaluating the switching life of a GaN power conversion device according to claim 2, characterized by:
the test conditions for the current stress-failure test were set as follows:
controlling the voltage between the drain electrode and the source electrode of the GaN power switch tube to keep the rated voltage value unchanged, and keeping the junction temperature to a preset value according to a preset gradient value;
the test conditions for the voltage stress-failure test were set as follows:
controlling the current between the drain electrode and the source electrode of the GaN power switch tube to keep the rated current unchanged, and keeping the junction temperature at a preset value according to a preset gradient value of the voltage;
the test conditions for the temperature stress-failure test were set as follows:
and keeping the rated current of the current between the drain electrode and the source electrode in the on state of the GaN power switch tube or keeping the rated voltage of the voltage between the drain electrode and the source electrode in the off state of the GaN power switch tube unchanged, and taking the value of the junction temperature according to a preset gradient.
9. The method for evaluating the switching life of a GaN power conversion device according to claim 1, wherein the multi-factor accelerated life model is:
AF＝K 1 ·f v (V DS )·f i (I DP )·f T (T)
wherein AF is an acceleration factor; f (f) v (V DS ) F is a voltage stress acceleration life model i (I DP ) F is a current stress acceleration life model T (T) is a temperature stress acceleration life model, V DS For the drain-to-source voltage of GaN power switch tube, I DP The drain current in the on state of the GaN power switch is T, the junction temperature of the GaN power switch tube is K 1 And accelerating the life model coefficient for multiple factors.
10. The method for evaluating the switching life of the GaN power conversion device according to claim 2, wherein the current stress-failure test, the temperature stress-failure test and the voltage stress-failure test are all performed by a test method using a number of truncations until all circuits fail, and the test is stopped.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310638623.0A CN116840670A (en) | 2023-05-31 | 2023-05-31 | Switch life assessment method of GaN power conversion device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310638623.0A CN116840670A (en) | 2023-05-31 | 2023-05-31 | Switch life assessment method of GaN power conversion device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116840670A true CN116840670A (en) | 2023-10-03 |
Family
ID=88171481
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310638623.0A Pending CN116840670A (en) | 2023-05-31 | 2023-05-31 | Switch life assessment method of GaN power conversion device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116840670A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117330951A (en) * | 2023-11-16 | 2024-01-02 | 华羿微电子股份有限公司 | SOA (service oriented architecture) test system and test method for MOS (metal oxide semiconductor) switch device |
-
2023
- 2023-05-31 CN CN202310638623.0A patent/CN116840670A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117330951A (en) * | 2023-11-16 | 2024-01-02 | 华羿微电子股份有限公司 | SOA (service oriented architecture) test system and test method for MOS (metal oxide semiconductor) switch device |
| CN117330951B (en) * | 2023-11-16 | 2024-03-22 | 华羿微电子股份有限公司 | SOA (service oriented architecture) test system and test method for MOS (metal oxide semiconductor) switch device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ye et al. | Online condition monitoring of power MOSFET gate oxide degradation based on miller platform voltage | |
| Puschkarsky et al. | Understanding BTI in SiC MOSFETs and its impact on circuit operation | |
| Milor et al. | Minimizing production test time to detect faults in analog circuits | |
| Saha et al. | Accelerated aging with electrical overstress and prognostics for power MOSFETs | |
| KR101478554B1 (en) | Method for producing overshoot voltage supplied at transistor and gate insulation degradation analysis used the same | |
| US6633177B1 (en) | Method of predicting lifetime of semiconductor integrated circuit and method for reliability testing of the circuit | |
| Duan et al. | Key issues and solutions for characterizing hot carrier aging of nanometer scale nMOSFETs | |
| Li et al. | A new SPICE reliability simulation method for deep submicrometer CMOS VLSI circuits | |
| CN116840670A (en) | Switch life assessment method of GaN power conversion device | |
| Gonzalez et al. | Bias temperature instability and condition monitoring in SiC power MOSFETs | |
| Paine | Scaling DC lifetests on GaN HEMT to RF conditions | |
| US7340360B1 (en) | Method for determining projected lifetime of semiconductor devices with analytical extension of stress voltage window by scaling of oxide thickness | |
| Nafria et al. | Breakdown of thin gate silicon dioxide films—A review | |
| Bilir et al. | A comparative study on power MOSFET reliability and failure modes | |
| Hunter | The statistical dependence of oxide failure rates on V/sub dd/and t/sub ox/variations, with applications to process design, circuit design, and end use | |
| Malobabic et al. | Gate oxide evaluation under very fast transmission line pulse (VFTLP) CDM-type stress | |
| US6825684B1 (en) | Hot carrier oxide qualification method | |
| Wang et al. | A reliability assessment system for power MOSFET using multi-parameters | |
| Gómez et al. | Deep Investigation on SiC MOSFET Degradation under Gate Switching Stress and Application Switching Stress | |
| Capasso et al. | A novel approach to analyze the reliability of GaN power HEMTs operating in a DC-DC Buck converter | |
| Sunter et al. | Quick Analyses for Improving Reliability and Functional Safety of Mixed-Signal ICs | |
| Efthymiou et al. | A methodology for projecting SiO2 thick gate oxide reliability on trench power MOSFETs and its application on MOSFETs VGS rating | |
| Sayed et al. | Dynamic remaining useful lifetime (RUL) estimation of power converters based on GaN power FETs | |
| Ren et al. | A method for health monitoring of power MOSFETs based on threshold voltage | |
| Nafria et al. | Characterization of SiO/sub 2/dielectric breakdown for reliability simulation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |