Switching loss measurement method and system based on total loss and switching frequency slope

Abstract

The application provides a switching loss measurement method and system based on total loss and switching frequency slope, wherein the method comprises the following steps: connecting a direct current source or a direct voltage source to both ends of a measured device by using a relay network for heating, and obtaining a thermal impedance model from a first reference point to a second reference point; controlling the measured device to run in a high-frequency switching state and flow through a square wave current with constant amplitude and duty cycle, measuring the temperature of the first reference point and the second reference point when a thermal steady state is reached, and calculating the total loss power by using the thermal impedance model from the first reference point to the second reference point; changing the switching frequency multiple times to repeat the experiment, obtaining a curve of the total loss power changing with the switching frequency, and fitting the curve, so that the curve intercept is the conduction loss under the given test condition, and the slope is the switching loss under the given test condition, and the switching loss measurement is completed. The application can reduce the cost of the measurement equipment, can test multiple characteristic parameters at the same time, and improves the test efficiency.

Description

Technical Field

[0001] This invention relates to the field of power semiconductor device switching loss measurement technology, specifically to a method and system for measuring switching losses based on the relationship between total loss and switching frequency slope. Background Technology

[0002] During high-frequency switching operation of power electronic converters, power semiconductor devices experience significant switching losses, leading to increased junction temperatures. Therefore, accurate understanding of the switching loss characteristics of devices is crucial throughout the entire converter lifecycle, including device selection, design and manufacturing, and operational status monitoring, to ensure the safe and reliable operation of power semiconductor devices. Currently, the switching loss characteristics of power semiconductor devices are primarily extracted through double-pulse testing.

[0003] Double-pulse testing applies voltage and current pulses to the device under test (DUT) using a dedicated experimental circuit, measures and integrates the voltage and current during the turn-on and turn-off processes to obtain the energy loss generated during the process. This method has two drawbacks: First, because the circuit structure and operating state of double-pulse testing are relatively simple and ideal, actual converters often employ more complex structures such as full-bridge or three-phase bridge converters and operate in high-frequency switching states, resulting in insufficient accuracy of the double-pulse test results. Second, double-pulse testing is limited by its measurement principle, requiring voltage and current sensors with high bandwidth and accuracy, thus increasing the testing cost. Summary of the Invention

[0004] To address the aforementioned shortcomings in the prior art, this invention provides a method and system for measuring switching losses based on total losses and the slope of the switching frequency.

[0005] According to one aspect of the present invention, a method for measuring switching losses based on total losses and switching frequency slope is provided, comprising:

[0006] A relay network is used to connect a DC current source or a DC voltage source to the two ends of the device under test for heating, thereby obtaining a thermal impedance model from the first reference point to the second reference point.

[0007] The device under test is controlled to operate in a high-frequency switching state, and a square wave current with constant amplitude and duty cycle flows through the device under test. The total power loss under the corresponding control mode is determined. When the device under test reaches thermal steady state, the temperatures of the first reference point and the second reference point are measured, and the total power loss is calculated using the thermal impedance model from the first reference point to the second reference point.

[0008] Keeping the voltage, junction temperature, current, and duty cycle of the device under test constant, the experiment is repeated by changing the switching frequency of the device under test multiple times. During this time, the average switching power loss of the device under test changes proportionally with the switching frequency, while the average conduction power loss remains constant. The total power loss of the device under test is the sum of the average switching power loss and the average conduction power loss. The average switching power loss is the product of the energy loss generated by a single switch and the switching frequency. The average conduction power loss is the product of the on-state voltage drop, current, and duty cycle.

[0009] Based on data obtained from repeated experiments, the switching frequency of the device under test is used as the independent variable and the total power loss is used as the dependent variable. The relationship between the two is fitted using a linear function. The intercept of the function is the average conduction power loss of the device under test, and the slope of the function is the energy loss generated by the device under test in a single switching operation. The switching loss measurement is thus completed.

[0010] According to another aspect of the present invention, a switching loss measurement system based on total loss and switching frequency slope is provided, comprising:

[0011] A multi-device thermal impedance model extraction module uses a relay network to connect a DC current source or a DC voltage source to both ends of the device under test for heating, thereby obtaining a thermal impedance model from a first reference point to a second reference point.

[0012] A constant current thermal steady-state operation module for switching is used to control the device under test to operate in a high-frequency switching state and to make the device under test flow with a square wave current with constant amplitude and duty cycle, and to determine the total power loss during operation under the corresponding control mode.

[0013] The module for measuring temperature rise and back-calculating total loss is used to measure the temperature of the first reference point and the second reference point after the device under test reaches thermal steady state, and to calculate the total power loss using the thermal impedance model from the first reference point to the second reference point.

[0014] A multi-switching frequency total loss calculation module is used to keep the voltage, junction temperature, current, and duty cycle of the device under test (DUT) constant while repeatedly changing the switching frequency of the DUT. During this process, the average switching power loss of the DUT changes proportionally to the switching frequency, while the average conduction power loss remains constant. The total power loss of the DUT is the sum of the average switching power loss and the average conduction power loss. The average switching power loss is the product of the energy loss generated during a single switch and the switching frequency. The average conduction power loss is the product of the on-state voltage drop, current, and duty cycle. Based on the data obtained from multiple repeated experiments, with the switching frequency of the DUT as the independent variable and the total power loss as the dependent variable, a linear function is used to fit the relationship between the two. The intercept of the function is the average conduction power loss of the DUT, and the slope of the function is the energy loss generated during a single switch of the DUT, thus completing the switching loss measurement.

[0015] By adopting the above technical solution, the present invention has at least one of the following beneficial effects compared with the prior art:

[0016] The switching loss measurement method and system based on total loss and switching frequency slope provided by this invention adopts a constant current thermal steady-state test state for switching, which is very close to the circuit parasitic parameters and device self-heating operation mode of actual converters, thus making the test results more accurate and more meaningful.

[0017] The switching loss measurement method and system based on total loss and switching frequency slope provided by this invention employs a method for back-deriving total loss by measuring temperature rise and a method for calculating total loss at multiple switching frequencies. This avoids high-bandwidth voltage and current integration, thereby significantly reducing the sensor cost required for testing.

[0018] The present invention provides a method and system for measuring switching losses based on total loss and switching frequency slope. The multi-device thermal impedance model extraction method, the method of measuring temperature rise to infer total loss, and the multi-switching frequency total loss calculation method adopted can extract thermal impedance characteristics and conduction loss characteristics in the same test circuit while extracting switching loss characteristics, which greatly improves the test efficiency compared with existing methods. Attached Figure Description

[0019] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0020] Figure 1 This is a flowchart illustrating the switching loss measurement method based on total loss and switching frequency slope in a preferred embodiment of the present invention.

[0021] Figure 2This is a schematic diagram of the components of a switching loss measurement system based on total loss and switching frequency slope in a preferred embodiment of the present invention.

[0022] Figure 3 This is a flowchart illustrating the workflow of a switching loss measurement method based on total loss and switching frequency slope in a specific application example of the present invention. Detailed Implementation

[0023] The embodiments of the present invention are described in detail below: These embodiments are implemented based on the technical solution of the present invention, and provide detailed implementation methods and specific operation processes. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention.

[0024] One embodiment of the present invention provides a method for measuring switching losses based on total losses and switching frequency slope. This method achieves centralized extraction of thermal impedance model, on-state voltage drop, and switching loss characteristics in the same test circuit and test process, which can reduce the cost of measurement equipment and can centrally test multiple characteristic parameters, thereby improving test efficiency.

[0025] Specifically, such as Figure 1 As shown, the switching loss measurement method based on total loss and switching frequency slope provided in this embodiment may include:

[0026] S1, use a relay network to connect a DC current source or a DC voltage source to the two ends of the device under test for heating, and obtain the thermal impedance model from the first reference point to the second reference point;

[0027] S2, control the device under test to operate in a high-frequency switching state, and make the device under test flow with a square wave current with constant amplitude and duty cycle, and determine the total power loss under the corresponding control mode; after the device under test reaches thermal steady state, measure the temperature of the first reference point and the second reference point, and calculate the total power loss using the thermal impedance model from the first reference point to the second reference point.

[0028] S3, keeping the voltage, junction temperature, current, and duty cycle of the device under test constant, repeatedly perform the previous step by changing the switching frequency of the device under test multiple times to repeat the experiment. At this time, the average switching power loss of the device under test changes proportionally with the switching frequency while the average conduction power loss remains constant. The total power loss of the device under test is the sum of the average switching power loss and the average conduction power loss. The average switching power loss is the product of the energy loss generated by a single switch and the switching frequency. The average conduction power loss is the product of the on-state voltage drop, current, and duty cycle.

[0029] S4. Based on the data obtained from repeated experiments, the switching frequency of the device under test is used as the independent variable and the total power loss is used as the dependent variable. The relationship between the two is fitted using a linear function. The intercept of the function is the average conduction power loss of the device under test, and the slope of the function is the power loss generated by the device under test in a single switching operation. The switching loss measurement is then completed.

[0030] In some preferred embodiments, the above-mentioned S1, in which the device under test is constructed by connecting the power semiconductor devices under test into a bridge circuit, may further include:

[0031] Two power semiconductor devices under test are connected in series to form a bridge arm, forming three nodes: the positive terminal of the bridge arm, the negative terminal of the bridge arm, and the midpoint of the bridge arm. The positive terminals of different bridge arms are connected to each other, and the negative terminals of different bridge arms are connected to each other, resulting in a bridge circuit composed of one or more bridge arms.

[0032] In some preferred embodiments, the above-mentioned S1, which uses a relay network to connect a DC current source to both ends of the device under test for heating, may further include:

[0033] The relay network consists of multiple independently controlled relays. The positive and negative terminals of the DC current source are connected to the positive, negative, and midpoint of each bridge arm of the circuit under test, and the relays are controlled to open and close. This allows the DC current source to be connected to one or more devices under test for heating as needed, or the DC voltage source to be connected to the positive and negative terminals of the bridge arm for heating as needed.

[0034] In some preferred embodiments, the above-mentioned S1, obtaining the thermal impedance model from the first reference point to the second reference point, may further include:

[0035] S11, keep the device under test constantly connected, connect the DC current source across the two ends of the device under test and pass a constant heating current;

[0036] S12. After the device under test reaches thermal steady state, measure the on-state voltage drop and heating current of the device under test at this time, and then cut off the constant heating current to enter the cooling process.

[0037] S13, during the cooling process, continuously measure the temperature of the first reference point and the second reference point of the device under test, flip the cooling curves of the first reference point temperature and the second reference point temperature, and shift the temperature at the beginning time to coincide, to obtain the heating process curve.

[0038] S14, divide the temperature difference between the first reference point and the second reference point in the inverted heating process curve by the product of the on-state voltage drop and the constant heating current to establish the transient thermal impedance model from the first reference point to the second reference point:

[0039] ;

[0040] In the formula, Z th ( t ) for the semiconductor device under test t The transient thermal impedance from the first reference point to the second reference point at any given moment. T r1 ( t () represents the time point in the heating process curve. t The first reference point temperature, T r2 ( t () represents the time point in the heating process curve. t The second reference point temperature, V ce This is the on-state voltage drop under thermal steady-state conditions. I heat This is the heating current;

[0041] S15, with t Increase Z th ( t The thermal resistance gradually increases and approaches a steady-state value, which is the steady-state thermal resistance from the first reference point to the second reference point. R th .

[0042] In some preferred embodiments, S2, which controls the device under test to operate in a high-frequency switching state and allows a square wave current with a constant amplitude and duty cycle to flow through the device under test, and determines the total power loss during operation under the corresponding control mode, may further include:

[0043] S21, connect a DC voltage source to the positive and negative terminals of the bridge arm, controlling the device under test (DUT) to operate in a high-frequency switching state and reach thermal steady state. At this time, a constant power loss is generated. The test current through the DUT is constant, and the duty cycle is constant, resulting in a constant average conduction power loss, expressed as:

[0044] ;

[0045] In the formula, P cond The average conduction loss power of the device under test. V ce This is the on-state voltage drop under thermal steady-state conditions. I test To test the current, D The duty cycle of the device under test;

[0046] S22, the energy loss generated by a single switch of the device under test is constant, and the switching frequency of the device under test is also constant, resulting in a constant average switching power loss, expressed as:

[0047] ;

[0048] In the formula, P sw The average switching loss power of the device under test. f sw The switching frequency of the device under test. E sw The energy loss generated by a single switching operation of the device under test;

[0049] S23, the total power loss is the sum of the average conduction power loss and the average switching power loss, expressed as:

[0050] ;

[0051] In the formula, P loss This represents the total power loss of the device under test.

[0052] In some preferred embodiments, the above-mentioned step S2, which involves measuring the temperature of the first reference point and the temperature of the second reference point after the device under test reaches thermal steady state, and calculating the total power loss using a thermal impedance model from the first reference point to the second reference point, may further include:

[0053] S24, After the device under test (DUT) reaches thermal steady state through high-frequency switching operation, the total power loss is calculated using the first and second reference point temperatures measured when the DUT reaches thermal steady state, as well as the thermal impedance model:

[0054] ;

[0055] In the formula, P loss The total power loss of the device under test. T r1 The temperature of the first reference point of the device under test in thermal steady state. T r2 This is the second reference point temperature under thermal steady state. R th The steady-state thermal resistance of the device under test from the first reference point to the second reference point is given.

[0056] In some preferred embodiments, S3 above, keeping the voltage, junction temperature, current, and duty cycle of the device under test constant, and repeatedly changing the switching frequency of the device under test, the experiment is repeated. In this case, the average switching loss power of the device under test changes proportionally to the switching frequency while the average conduction loss power remains constant. It may further include:

[0057] S31: Keep the test voltage, test current, duty cycle and junction temperature constant, change the switching frequency of the device under test, so that the average switching power of the device under test changes, while the average conduction power remains unchanged.

[0058] S32, after completing a total power loss measurement, the switching frequency is changed to repeat the process of calculating the total power loss, so as to obtain the total power loss of the device under test at different switching frequencies.

[0059] S33, expressing the total power loss as a linear function of the switching frequency is:

[0060] ;

[0061] In the formula, the switching frequency is... f sw The total power loss is the independent variable in the functional relationship. P loss The dependent variable in the functional relationship is the average conduction loss power of the device under test, which is a constant of the intercept term, derived from the conduction voltage drop under thermal steady state. V ce Test current I test Duty cycle D The product of these values ​​gives the energy loss generated by a single switching operation of the device under test. E sw The slope term constant;

[0062] S34. Based on the results obtained from multiple calculations of total power loss, the intercept of a linear function is fitted to obtain the average conduction power loss of the device under test. P cond The on-state voltage drop of the device under test is calculated by combining the duty cycle and the test current:

[0063] ;

[0064] S35. Based on the results of multiple calculations of total power loss, the slope of a linear function is fitted to obtain the energy loss generated by a single switching operation of the device under test. E sw .

[0065] In some preferred embodiments, the above-mentioned S31 may further include:

[0066] When the switching frequency is changed, the total power loss of the device under test changes. At this time, the heating or cooling power of the external heat dissipation device is adjusted so that the junction temperature condition of the device under test remains unchanged when thermal steady state is reached each time.

[0067] In some preferred embodiments, S1 and S2 may further include:

[0068] The first reference point can be the junction reference point or any reference point on the hot path. When the junction reference point is used, the temperature of the first reference point is the junction temperature. The second reference point can be any reference point on the hot path.

[0069] In some preferred embodiments, the measurement of the temperatures at the first reference point and the second reference point in step S2 may further include:

[0070] The first reference point temperature is the junction temperature or case temperature of the device under test, or other reference point temperatures that vary significantly with the device's heat generation power.

[0071] The second reference point temperature is the case temperature of the device under test, the heat sink temperature, the ambient temperature, or the temperature of the integrated resistance temperature detector (RTD) within the power semiconductor module.

[0072] The junction temperature of the device is measured using temperature-sensitive electrical parameters, fiber optic sensors, or infrared thermometers.

[0073] Thermocouples, thermistors, fiber optic sensors, or infrared thermometers are used to measure the case temperature, heat sink temperature, ambient temperature, or the temperature of the resistance temperature detector integrated within a power semiconductor module.

[0074] Based on the same inventive concept, an embodiment of the present invention also provides a switching loss measurement system based on total loss and switching frequency slope.

[0075] Specifically, such as Figure 2 As shown, the switching loss measurement system based on total loss and switching frequency slope provided in this embodiment may include:

[0076] A multi-device thermal impedance model extraction module uses a relay network to connect a DC current source or a DC voltage source to the two ends of the device under test for heating, and obtains the thermal impedance model from the first reference point to the second reference point.

[0077] The constant current thermal steady-state operation module is used to control the device under test to operate in a high-frequency switching state and to make the device under test flow with a square wave current with constant amplitude and duty cycle, and to determine the total power loss during operation under the corresponding control mode.

[0078] The module for measuring temperature rise and back-calculating total loss is used to measure the temperature of the first reference point and the second reference point after the device under test reaches thermal steady state, and to calculate the total power loss using the thermal impedance model from the first reference point to the second reference point.

[0079] A multi-switching frequency total loss calculation module is used to keep the voltage, junction temperature, current, and duty cycle of the device under test (DUT) constant while repeatedly changing the switching frequency of the DUT. During this process, the average switching power loss of the DUT changes proportionally to the switching frequency, while the average conduction power loss remains constant. The total power loss of the DUT is the sum of the average switching power loss and the average conduction power loss. The average switching power loss is the product of the energy loss generated during a single switch and the switching frequency. The average conduction power loss is the product of the on-state voltage drop, current, and duty cycle. Based on the data obtained from multiple repeated experiments, with the switching frequency of the DUT as the independent variable and the total power loss as the dependent variable, a linear function is used to fit the relationship between the two. The intercept of the function is the average conduction power loss of the DUT, and the slope of the function is the energy loss generated during a single switch of the DUT, thus completing the switching loss measurement.

[0080] It should be noted that the steps in the method provided by the present invention can be implemented using corresponding modules, devices, units, etc. in the system. Those skilled in the art can refer to the technical solution of the method to realize the composition of the system. That is, the embodiments in the method can be understood as preferred examples for building the system, and will not be elaborated here.

[0081] The technical solution provided by the above embodiments of the present invention will be further described in detail below with reference to a specific application example.

[0082] like Figure 3 As shown, the switching loss measurement method based on total loss and switching frequency slope involved in this specific application example includes the following steps:

[0083] Step 1, Extraction of thermal impedance model of multiple devices: The power semiconductor devices under test are connected into a bridge circuit. A DC current source is connected to both ends of any device under test in the bridge circuit using a relay network for heating. The junction temperature of the device under test and the reference point temperature are measured during the process, and then the thermal impedance model from junction to reference point is obtained.

[0084] Step 2, Constant Current Thermal Steady-State Operation of Switch: Apply a constant test voltage to the bus of the bridge circuit, and control the device under test to operate in a high-frequency switching state through the drive signal, so that the current flowing through the device under test is a square wave current with constant amplitude and constant duty cycle. The maximum value of the square wave current is the set test current, and the minimum value is zero.

[0085] Step 3, Measure temperature rise to deduce total loss: After the device under test (DUT) reaches thermal steady state under high-frequency switching conditions, calculate the total power loss of the DUT under thermal steady state using the temperature rise and thermal impedance model.

[0086] ;

[0087] in, Ploss The total power loss of the device under test. T j This refers to the junction temperature of the device under test in its thermal steady state. T r This is the reference temperature at thermal steady state. R jr The steady-state thermal resistance of the device under test junction to the reference point is given.

[0088] Step 4, Multi-Switching Frequency Total Loss Data Scan: Repeat the experiment by changing the switching frequency and reference point temperature of the device under test, so that the total power loss changes while the junction temperature and average conduction power loss remain unchanged, and record multiple sets of switching frequency and total power loss data of the device under test under test voltage and test current.

[0089] Step 5, Fitting the Total Loss-Switching Frequency Relationship: Using the switching frequency of the device under test as the independent variable and the total power loss as the dependent variable, fit a linear function relationship as follows:

[0090] ;

[0091] In the functional relationship, the independent variable is the switching frequency. f sw The dependent variable is the total power loss. P loss The slope term constant is the energy loss generated by a single switching operation of the device under test. E sw The intercept term constant is the on-state voltage drop of the device under test in thermal steady state. V ce Test current I test Duty cycle D The product of total loss and switching frequency. By fitting the slope and intercept of the total loss-switching frequency relationship function, the switching loss and on-state voltage drop characteristics of the device under test are indirectly measured.

[0092] In this specific application example, the method for obtaining the thermal impedance model from the junction to the reference point includes:

[0093] To keep the device under test constantly connected, connect a DC current source across its two ends and pass a constant heating current through it;

[0094] After the device under test reaches thermal steady state, the on-state voltage drop and heating current of the device under test are measured. Then the constant heating current is cut off and the cooling process begins.

[0095] During the cooling process, the junction temperature and reference point temperature of the device under test are continuously measured. The cooling curves of the junction temperature and reference point temperature are flipped and the initial temperature is shifted to coincide, thus obtaining the heating process curve.

[0096] The transient thermal impedance from the junction to the reference point is calculated by dividing the junction-to-reference point temperature difference in the heating process curve obtained by inversion by the product of the on-state voltage drop and the constant heating current.

[0097] ;

[0098] in, Z jr ( t ) for the device under test t The transient thermal impedance from the junction to the reference point at any given time. T j ( t () represents the time point in the heating process curve. t The junction temperature, T r ( t () represents the time point in the heating process curve. t The reference point temperature, V ce This is the on-state voltage drop under thermal steady-state conditions. I heat This is the heating current;

[0099] along with t The increase, Z jr ( t The thermal resistance gradually increases and approaches the steady-state value, which is the steady-state thermal resistance from the junction to the reference point. R jr ;

[0100] Furthermore, the junction temperature and reference point temperature in the transient thermal impedance formula can be replaced with the temperature between any two reference points on the thermal path, and the transient thermal impedance and steady-state thermal resistance between the two reference points can be obtained.

[0101] In this specific application example, the devices under test are connected in a bridge circuit, and a DC current source is connected to either end of the device under test through a relay network for heating, including:

[0102] The devices under test are connected to form a bridge circuit, which consists of one or more bridge arms. Each bridge arm consists of two devices under test connected in series, forming three nodes: the positive terminal of the bridge arm, the negative terminal of the bridge arm, and the midpoint of the bridge arm. The positive terminals of different bridge arms are connected to each other, and the negative terminals of different bridge arms are connected to each other.

[0103] The relay network consists of multiple independently controlled relays. The positive and negative terminals of the DC current source are connected to the positive, negative, and midpoint nodes of each arm of the bridge circuit under test. The on / off state of the relays is controlled by programming, so that the DC current source can be connected to one or more devices under test as needed.

[0104] In this specific application example, when measuring switching losses, the device under test operates in a high-frequency switching state and reaches a thermal steady state, at which point it generates constant power loss, including:

[0105] Since the test current and duty cycle of the device under test are constant, the resulting average conduction loss power is constant, which can be expressed as:

[0106] ;

[0107] in, P cond The average conduction loss power of the device under test. V ce This is the on-state voltage drop under thermal steady-state conditions. I test To test the current, D The duty cycle of the device under test;

[0108] Since the voltage and current of the device under test are constant during each switching operation, the energy loss generated in a single switching operation is constant. Furthermore, the switching frequency of the device under test is also constant, resulting in a constant average switching power loss, which can be expressed as:

[0109] ;

[0110] in, P sw The average switching loss power of the device under test. f sw The switching frequency of the device under test. E sw The energy loss generated by a single switching operation of the device under test;

[0111] The total power loss is the sum of the average conduction power loss and the average switching power loss, and therefore remains constant, which can be expressed as:

[0112] ;

[0113] in, P loss This represents the total power loss of the device under test.

[0114] In this specific application example, after the device under test (DUT) reaches thermal steady state through high-frequency switching operation, the total power loss is calculated using temperature measurements and a thermal impedance model of the DUT at that point. The calculation formula is as follows:

[0115] ;

[0116] in, P loss The total power loss of the device under test. Tj This refers to the junction temperature of the device under test in its thermal steady state. T r This is the reference temperature at thermal steady state. R jr The steady-state thermal resistance of the device under test junction to the reference point;

[0117] Furthermore, the total power loss can be calculated using the temperature difference and thermal resistance model of two other points on the same thermal path. The calculation formula is as follows:

[0118] ;

[0119] in, T r1 and T r2 These are the temperatures of two different reference points in the same thermal path during thermal steady state. R r1r2 This represents the steady-state thermal resistance between two different reference points.

[0120] In this specific application example, the junction temperature and reference point temperature of the device under test are measured when the device reaches thermal steady state. The methods include, but are not limited to:

[0121] Junction temperature is measured using methods such as temperature-sensitive electrical parameters, fiber optic sensors, or infrared thermometers.

[0122] The reference point temperature is measured using the case temperature of the device under test, the heat sink temperature, the ambient temperature, or the temperature of the resistance temperature detector integrated in the power semiconductor module. The reference point temperature is measured using methods such as thermocouples, thermistors, fiber optic sensors, or infrared thermometers.

[0123] In this specific application example, the experiment was repeated by changing the switching frequency of the device under test, and multiple sets of switching frequency and total power loss data were tested and recorded. Specifically:

[0124] Keep the test voltage, test current, duty cycle, and junction temperature constant, and change the switching frequency of the device under test to change the average switching power loss of the device under test, while keeping the average conduction power loss constant.

[0125] After completing one measurement of total power loss, the experiment was repeated multiple times by changing the switching frequency. The total power loss of the device under test at different switching frequencies was obtained by temperature measurement and calculation.

[0126] In this specific application example, the total power loss of the device under test changes after the switching frequency is changed. At this time, it is necessary to adjust the heating or cooling power of the external heat dissipation device to ensure that the junction temperature of the device under test remains unchanged when the thermal steady state is reached in each repeated experiment.

[0127] In this specific application example, multiple switching frequency values ​​for the test need to be preset before testing. During the total loss data scanning phase of multiple switching frequencies, the judgment condition is set as "whether all preset switching frequencies have been tested". If yes, the preset values ​​have been traversed, and the total loss-switching frequency relationship fitting is further performed. If not, the experiment process is restarted. In order to fit a linear function curve, the number of selected switching frequencies should be greater than or equal to two. The more switching frequencies selected, the smaller the random error of the test will be, but the longer the test time will be.

[0128] In this specific application example, the total power loss is expressed as a linear function of the switching frequency, with the following formula:

[0129] ;

[0130] Among them, switching frequency f sw The total power loss is the independent variable in the functional relationship. P loss The dependent variable in the functional relationship is the average conduction loss power of the device under test, which is a constant of the intercept term, derived from the conduction voltage drop under thermal steady state. V ce Test current I test Duty cycle D The product of these values ​​gives the energy loss generated by a single switching operation of the device under test. E sw The slope term constant;

[0131] Based on the results obtained from repeated experiments, the average conduction loss power of the device under test can be obtained by fitting the intercept of a linear function. P cond Furthermore, by combining the duty cycle and the test current, the on-state voltage drop of the device under test can be calculated. The specific calculation formula is as follows:

[0132] ;

[0133] Based on the results obtained from repeated experiments, the energy loss generated by a single switching operation of the device under test can be obtained by fitting the slope of a linear function. E sw .

[0134] The switching loss measurement method and system based on total loss and switching frequency slope provided in the above embodiments of the present invention utilizes the relationship between total loss and switching frequency slope to connect the power semiconductor device under test (DUT) into a bridge circuit. First, a DC current source is connected to both ends of the DUT using a relay network for heating, and the junction-to-reference point thermal impedance model is extracted. Then, the DUT is operated in a high-frequency switching state, with a square wave current of constant amplitude and duty cycle flowing through it. After reaching thermal steady state, the junction temperature and reference point temperature are measured, and the total power loss is calculated using the junction-to-reference point thermal impedance model. The experiment is repeated with multiple changes in the switching frequency, and a curve of total power loss versus switching frequency is plotted and fitted using a linear function. The intercept of the obtained curve represents the conduction loss under given test conditions, and the slope represents the switching loss under given test conditions. This method can reduce the cost of measurement equipment and can centrally test multiple characteristic parameters, improving testing efficiency.

[0135] Any matters not covered in the above embodiments of the present invention are well-known in the art.

[0136] The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various modifications or variations within the scope of the claims, which do not affect the essence of the present invention.

Claims

1. A method of measuring switching losses based on total losses and switching frequency slope, characterized by, The method comprises the following steps: a DC current source or a DC voltage source is connected to both ends of the device under test by a relay network for heating, and a thermal impedance model from the first reference point to the second reference point is obtained; the device under test is controlled to operate in a high-frequency switching state, and a square wave current with constant amplitude and duty cycle is passed through the device under test to determine the total loss power when the device under test operates in the corresponding control mode; when the device under test reaches a thermal steady state, the temperatures of the first reference point and the second reference point are measured, and the total loss power is calculated by using the thermal impedance model from the first reference point to the second reference point; the voltage, junction temperature, current and duty cycle of the device under test are kept unchanged, the switching frequency of the device under test is changed multiple times, and the experiment is repeated; at this time, the average switching loss power of the device under test is proportional to the switching frequency, and the average conduction loss power is constant; wherein the total loss power of the device under test is the sum of the average switching loss power and the average conduction loss power, the average switching loss power is the product of the loss energy generated by a single switch and the switching frequency, and the average conduction loss power is the product of the conduction voltage drop, the current and the duty cycle; based on the data obtained from multiple repeated experiments, the switching frequency of the device under test is taken as the independent variable, and the total loss power is taken as the dependent variable; a first-order function is used to fit the relationship between the two, the intercept of the function is the average conduction loss power of the device under test, and the slope of the function is the loss energy generated by a single switch of the device under test, thereby completing the switching loss measurement.

2. The method of claim 1, wherein the slope of the total loss versus switching frequency is determined by fitting a line to the total loss versus switching frequency data. The device under test is obtained by connecting the power semiconductor devices to be tested into a bridge circuit, and comprises: two power semiconductor devices to be tested are connected in series to form a bridge arm, forming three nodes of a bridge arm positive electrode, a bridge arm negative electrode and a bridge arm midpoint; the bridge arm positive electrodes of different bridge arms are connected to each other; the bridge arm negative electrodes of different bridge arms are connected to each other; and a bridge circuit is obtained by connecting one or more bridge arms. The relay network is composed of multiple independently controlled relays, and the positive electrode and the negative electrode of the DC current source are connected to the bridge arm positive electrode, the bridge arm negative electrode and the bridge arm midpoint of each bridge arm of the device under test, and the on-off of the relays is controlled, so that the DC current source is connected to both ends of one or more devices under test according to the test requirements, or the DC voltage source is connected to both ends of the bridge arm positive electrode and the bridge arm negative electrode according to the test requirements, for heating. The thermal impedance model from the first reference point to the second reference point is obtained by:

3. The switching loss measurement method based on total loss and switching frequency slope according to claim 1, characterized in that, keeping the device under test always on, connecting a DC current source to both ends of the device under test and passing a constant heating current; after the device under test reaches a thermal steady state, the conduction voltage drop and the heating current of the device under test at this time are measured, and then the constant heating current is removed and a cooling process is started; during the cooling process, the temperatures of the first reference point and the second reference point of the device under test are continuously measured, the cooling curves of the first reference point and the second reference point are flipped, and the temperatures at the starting time are translated to coincide, to obtain a heating process curve. ​ The temperature difference between the first reference point and the second reference point in the obtained temperature rising curve of the flip is divided by the product of the on-voltage drop and the constant heating current to obtain a transient thermal impedance model from the first reference point to the second reference point, and the transient thermal impedance model is: ； In the formula, Z th ( t ) for the semiconductor device under test t The transient thermal impedance from the first reference point to the second reference point at any given moment. T r1 ( t () represents the time point in the heating process curve. t The first reference point temperature, T r2 ( t () represents the time point in the heating process curve. t The second reference point temperature, V ce This is the on-state voltage drop under thermal steady-state conditions. I heat This is the heating current; With t increasing, Z th ( t ) gradually increases and approaches a steady-state value, which is the steady-state thermal resistance from the first reference point to the second reference point R th .

4. The switching loss measurement method based on total loss and switching frequency slope according to claim 1, characterized in that, The control of the measured device operating in a high-frequency switching state, and the measured device flowing through a square wave current with constant amplitude and constant duty cycle, determines the total loss power when operating in the corresponding control mode, including: The direct current voltage source is connected between the positive electrode of the bridge arm and the negative electrode of the bridge arm, the measured device is controlled to operate in a high-frequency switching state and reaches a thermal steady state, at this time a constant loss power is generated, the test current of the measured device is constant and the duty cycle is constant, the average on-loss power generated is constant, which is represented as: ； wherein P cond Pon is the average on-state loss power of the device under test, V ce Von is the on-state voltage drop at thermal steady state, I test Itest is the test current, D Duty is the duty cycle of the device under test; The loss energy generated by the measured device in a single switching is constant, and the switching frequency of the measured device is also constant, so the average switching loss power generated is constant, which is represented as: ； wherein P sw Pswavg is the average switching loss power of the device under test, f sw fsw is the switching frequency of the device under test, E sw Esw is the energy dissipated per switching of the device under test. The total loss power is the sum of the average on-loss power and the average switching loss power, which is represented as: ； In the formula, P loss Ptotai is the total power dissipated by the device under test.

5. The method of claim 1, wherein, The first reference point temperature and the second reference point temperature are measured after the measured device reaches the thermal steady state, and the total loss power is calculated by using the thermal impedance model from the first reference point to the second reference point, including: After the measured device operates in a high-frequency switching state and reaches a thermal steady state, the total loss power is calculated by using the first reference point temperature and the second reference point temperature measured when the measured device reaches the thermal steady state and the thermal impedance model, which is: ； wherein, P loss Ptot is the total power dissipated by the device under test, T r1 Trefi is the first reference point temperature of the device under test at thermal steady state, T r2 Tref2 is the second reference point temperature at thermal steady state, R th Rth is the steady state thermal resistance from the first reference point to the second reference point of the device under test.

6. The method of claim 1, wherein, The voltage, junction temperature, current and duty cycle of the measured device are kept unchanged, the switching frequency of the measured device is changed repeatedly in the repeated experiment, at this time the average switching loss power of the measured device changes in proportion to the switching frequency, and the average on-loss power remains unchanged, including: The test voltage, test current, duty cycle and junction temperature conditions are kept unchanged, the switching frequency of the measured device is changed, so that the average switching loss power of the measured device changes, while the average on-loss power remains unchanged; after completing the measurement of the total loss power once, the switching frequency is changed to repeat the process of calculating the total loss power, so as to obtain the total loss power of the measured device under different switching frequencies; The total loss power is represented as a first-order function of the switching frequency as: ； In the formula, the switching frequency f sw is the independent variable in the function relationship, the total loss power P loss is the dependent variable in the function relationship, the average conduction loss power of the measured device is the intercept term constant, and the conduction voltage drop at thermal steady state V ce , the test current I test and the duty cycle D are multiplied to obtain the loss energy generated by the measured device in a single switching E sw is the slope term constant; According to the results of the total loss power obtained by multiple calculations, the intercept of the first function is fitted to obtain the average conduction loss power of the measured device P cond And the conduction voltage drop of the measured device is calculated by combining the duty cycle with the test current: ； According to the results of the total loss power obtained by multiple calculations, the slope of the linear function is fitted to obtain the loss energy generated by a single switch of the measured device E sw .

7. The method of claim 6, wherein the slope of the total loss versus switching frequency is determined by fitting a line to the total loss versus switching frequency data. It also includes: When the switching frequency is changed, the total power loss of the measured device changes, at this time the heating or cooling power of the external heat dissipation device is adjusted, so that the junction temperature condition of the measured device is unchanged when reaching the thermal steady state each time.

8. The total loss versus switching frequency slope based switching loss measurement method of any of claims 1-7, wherein, The first reference point adopts a junction reference point or any reference point on a thermal path, when the junction reference point is adopted, the first reference point temperature adopts a junction temperature accordingly; the second reference point adopts any reference point on a thermal path.

9. The method of claim 1, wherein, The first reference point temperature and the second reference point temperature are measured, including: The first reference point temperature is the junction temperature or the shell temperature of the measured device, or the temperature of other reference points which change significantly with the heating power of the device; The second reference point temperature is the shell temperature of the measured device, the temperature of the heat sink, the ambient temperature or the temperature of the integrated thermal resistance in the power semiconductor module; The junction temperature of the device is measured by using a temperature-sensitive electrical parameter, an optical fiber sensor or an infrared thermometer; The temperature of the device under test, the temperature of the heat sink, the ambient temperature or the temperature of the integrated thermal resistance in the power semiconductor module is measured by using a thermocouple, a thermistor, an optical fiber sensor or an infrared temperature measuring instrument.

10. A total loss based switching loss measurement system with a slope of switching frequency, characterized by, The method comprises the following steps: a multi-device thermal impedance model extraction module, which is used to heat the device under test by connecting a direct current source or a direct voltage source to both ends of the device under test through a relay network, and to obtain a thermal impedance model from a first reference point to a second reference point; a switch constant current thermal steady state operation module, which is used to control the device under test to operate in a high-frequency switching state and to make the device under test flow through a square wave current with constant amplitude and duty cycle, and to determine the total loss power when the device under test operates in the corresponding control mode; a measurement temperature rise backstepping total loss module, which is used to measure the temperature of the first reference point and the second reference point after the device under test reaches a thermal steady state, and to calculate the total loss power by using the thermal impedance model from the first reference point to the second reference point; a multi-switching frequency total loss calculation module, which is used to keep the voltage, junction temperature, current and duty cycle of the device under test unchanged, to change the switching frequency of the device under test multiple times for repeated experiments, at this time, the average switching loss power of the device under test varies in direct proportion to the switching frequency, while the average conduction loss power remains unchanged; wherein the total loss power of the device under test is the sum of the average switching loss power and the average conduction loss power, the average switching loss power is the product of the loss energy generated by a single switch and the switching frequency, and the average conduction loss power is the product of the conduction voltage drop, the current and the duty cycle; according to the data obtained by multiple repeated experiments, taking the switching frequency of the device under test as the independent variable and the total loss power as the dependent variable, a first order function is used to fit the relationship between them, the intercept of the function is the average conduction loss power of the device under test, and the slope of the function is the loss energy generated by a single switch of the device under test, thereby completing the switching loss measurement.

Images