WO2020087928A1

WO2020087928A1 - Power cycle experiment device for automotive-grade igbt multi-junction temperature difference control

Info

Publication number: WO2020087928A1
Application number: PCT/CN2019/090606
Authority: WO
Inventors: 安彤; 赵静毅; 秦飞; 别晓锐; 袁雪泉
Original assignee: 北京工业大学
Priority date: 2018-10-28
Filing date: 2019-06-11
Publication date: 2020-05-07
Also published as: CN109444705B; CN109444705A

Abstract

A power cycle experiment device for automotive-grade IGBT multi-junction temperature difference control, comprising multiple IGBT module units as control modules and modules to be tested, a main circuit system for constituting a test loop, a small-current test system for junction temperature calculation, a water cooling system comprising multiple independent working positions, a control system for configuring IGBT drive and on/off of the water cooling system, and a data acquisition system for monitoring electrical and thermal parameters. The main circuit system adopts a specific series-parallel circuit to connect IGBT modules to be tested. Load currents of multiple IGBT modules to be tested in the same main circuit are made different. The data acquisition system automatically acquires the collector-emitter voltage V _ce of the IGBT module to be tested at a small current, and substitutes same into a junction temperature calculation formula to calculate the junction temperature T _j in real time. The experiment device simultaneously performs power cycle experiments with different junction temperature differences on multiple IGBT modules in one power cycle experiment, thereby meeting the diversity of experiments and improving the work efficiency.

Description

Automobile IGBT multi-junction temperature difference control power cycle experiment device

Technical field

The invention relates to a power cycle experimental device, which belongs to the field of experimental devices, in particular to a power cycle experimental device for automotive-grade IGBT multi-junction temperature difference control.

Background technique

Insulated gate bipolar transistor (IGBT) is the core device of energy conversion and transmission, which is widely used in aerospace, wind power, rail transit, electric vehicles and other fields. In practical applications, the power of the IGBT module changes, causing real-time changes in power loss, causing the junction temperature of the module to fluctuate widely. As the output power is further improved, the IGBT module will have serious overheating problems, the reliability of the IGBT module is reduced, and the aging failure occurs, which eventually leads to the failure of the entire power conversion system. Therefore, the reliability analysis of IGBT modules is crucial for the reliable operation of high-power converter devices.

To analyze the reliability of IGBT modules, accelerated aging experiments are required on IGBT modules in order to obtain more sample data in a shorter time. The power cycle experiment is an accelerated aging experiment commonly used at present. When conducting power cycle experiments, existing experimental methods can only perform power cycle experiments on IGBT modules at one experimental condition at a time. If you need to carry out a variety of power cycle experiments under different conditions, you need to conduct multiple experiments according to the experimental conditions, the experimental efficiency is low.

Therefore, a new power cycle experiment device needs to be proposed, which can perform power cycle experiments on multiple IGBT modules simultaneously. In addition, the experimental conditions such as current and junction temperature of multiple modules are different, which is convenient for comparison between multiple modules. This method greatly reduces the experimental time and improves the experimental efficiency.

Summary of the invention

In view of this, the present invention provides an automotive-grade IGBT multi-junction temperature difference control power cycle experiment device, which can simultaneously perform power cycle experiments with different junction temperature differences on multiple IGBT modules, which not only improves work efficiency, but also has more diverse experimental conditions.

The invention provides a power cycle experiment device for automotive-grade IGBT multi-junction temperature difference control. The device includes: an IGBT module, a main circuit system, a small current test system, a water-cooled heat dissipation system, a control system, and a data acquisition system. The IGBT module is placed on the water-cooled heat dissipation system, and the main circuit system and the small current test system are connected to the IGBT module respectively. The control system controls the gate of the IGBT1 module and the solenoid valve of the water cooling system. The data acquisition system monitors the data of the current I _{c of the} main circuit, the emitter voltage V _ce of the IGBT module to be tested, and the shell temperature T _{c of} the IGBT module to be tested in real time. And the V _ce data under the test current is automatically brought into the K curve, and the junction temperature T _{j of the} IGBT module is calculated in real time.

The IGBT module as shown in FIG. 1 includes five different types of IGBT modules, namely IGBT1, IGBT2, IGBT3, IGBT4 and IGBT5. The IGBT1 module with large rated power is used as the switch of the main circuit, and the large rated power is selected to ensure the reliability of the switch. IGBT2, IGBT3, IGBT4, IGBT5 of the same model with small rated power are used as the device under test to ensure the comparability of the experiment.

The main circuit system is mainly composed of power supply V1, resistance R1, IGBT1, IGBT2, IGBT3, IGBT4, IGBT5, as shown in Figure 1. The terminal 1 of the power supply V1 is connected to the 2 terminal of the resistor R1, and the 3 terminal of the resistor R1 is connected to the 4 terminal of the IGBT1. The two output terminals of IGBT2 and IGBT3 in parallel, one end is connected to the 5 end of IGBT1, the other end is connected to the 10 end of IGBT4, the 11 end of IGBT4 is connected to the 12 end of IGBT5, the 13 end of IGBT5 is connected to the 18 end of power supply V1 . In this circuit connection method, multiple IGBT modules to be tested use different series-parallel connection circuits to realize that multiple IGBT modules in the same main circuit have different load currents. The current flowing through IGBT2 and IGBT3 is half of the current in the main circuit, and the current flowing through IGBT4 and IGBT5 is equal to the current in the main circuit.

The calculation formula of the conduction power consumption of the IGBT module is

Where D is the duty cycle of the switching time, I _c is the load current of the IGBT module, V _ce (I _c ) is the collector-emitter voltage of the IGBT module, and V _{ce_25 °} C is the collector-emitter voltage of the IGBT module at 25 ° C , R _{ce_25 ℃} is the on-resistance of IGBT module at 25 ℃, K _V is the temperature coefficient of voltage, and K _r is the temperature coefficient of resistance.

The calculation formula of the junction temperature T _{j of the} IGBT module is

Among them, R _th is the thermal resistance of the IGBT, according to the data manual of the IGBT, the thermal resistance of the module under different experimental conditions is obtained, and T _c is the shell temperature of the IGBT.

It can be seen from formula (2) that the junction temperature of the IGBT is related to the power consumption of the IGBT and the shell temperature T _c of the water-cooled plate in the water-cooled heat dissipation system.

The small current test system is mainly composed of power supply V2 and diode D1, as shown in FIG. 1. The positive output 16 of the power supply V2 is connected to the 15 end of the diode D1, and the 14 of the diode D1 is connected to 5 of the IGBT1. The other output terminal 17 of the power supply is connected to the 13 terminal of the IGBT5.

The low current test system provides test current for four IGBT modules under test at the same time. Among them, the test current flowing into IGBT2 and IGBT3 is half of the test current of IGBT4 and IGBT5.

Further, the low current test system according to the function relationship curve between T _j and V _ce under different small currents in the thermostat experiment is shown in FIG. 2. Get the expression of T _j of IGBT at 50mA low current

Expression of T _j of IGBT at 100mA low current

Among them, the T _{j of} IGBT2 and IGBT3 is calculated using the T _j expression under the test current of 50 mA. The T _{j of} IGBT4 and IGBT5 is calculated using the T _j expression at 100 mA test current.

The water cooling system as shown in Figure 3. It can provide cooling liquid for multiple IGBT modules to dissipate heat. The cooling fluid (fluid) performs forced convection heat transfer in a circular pipe, and its convection heat transfer coefficient ɑ is

Where λ is the thermal conductivity, d is the pipe diameter, u is the liquid flow rate, ρ is the liquid density, μ is the liquid viscosity, c _p is the specific heat capacity, n is 0.4 when the fluid is heated, and 0.3 for cooling.

According to the convection heat transfer coefficient formula, it can be determined that when other parameters are constant, α is proportional to the 0.8 power of u, indicating that increasing the flow velocity u is beneficial to increase α.

When other parameters are fixed, α is inversely proportional to the 0.2 power of d, indicating that reducing the pipe diameter d is beneficial to increase α.

The formula of heat dissipation Q of water cooling system is

Q ＝ a · A · (T _c -T _w ) (6)

Where Q is the heat dissipation of the water-cooled plate, unit W; a is the convection heat transfer coefficient, unit W / (m ^2. ℃); A is the heat conduction area, unit m ² ; T _c is the temperature of the water-cooled plate (IGBT case temperature ), Which is the case temperature of the IGBT, in ° C; T _w is the temperature of the coolant, in ° C.

According to the heat dissipation formula of the water cooling system, when other parameters are fixed, Q is proportional to a. Therefore, the convection heat transfer coefficient a is adjusted by changing the pipe diameter d of the water cooling plate and the liquid flow rate u in the water cooling heat dissipation system, thereby controlling the heat dissipation of the water cooling heat dissipation system, and thereby controlling the IGBT case temperature T _c .

Further, the water cooling system includes five water cooling stations. Each station is equipped with an independent solenoid valve, which can independently control the on and off of each water-cooled station solenoid valve according to the heat dissipation requirements of the IGBT module, and then control the on and off time of the coolant.

The control system is controlled by a single-chip microcomputer, which can control the on-off of the switch module IGBT1 and the solenoid valve of the water cooling system.

The data collection system can simultaneously monitor the current I _{c of the} main circuit, the emitter voltage V _ce of the IGBT module to be tested and the shell temperature T _{c of} the IGBT module to be tested, and store the collected data in a computer in real time.

Further, the data collection system can automatically extract the data of the emitter voltage V _ce of the IGBT module to be tested under low current, and calculate the T _j change of the IGBT module according to the K curve of the IGBT module.

BRIEF DESCRIPTION

1 is a circuit diagram of a power cycle experiment device provided by the present invention;

2 is a graph corresponding to the IGBT collector emitter voltage V _ce and junction temperature T _j provided by the present invention;

Figure 3 is a water-cooled heat dissipation station diagram provided by the present invention;

4 is a flow chart of the power cycle experiment provided by the present invention;

FIG. 5 is a diagram for determining the power cycle experiment parameters provided by the present invention;

6 is a timing diagram of the power cycle experiment provided by the present invention;

detailed description

The specific implementation steps of the present invention will be described in detail below in conjunction with the accompanying drawings of the specification.

The invention provides a power cycle experiment device for automotive grade IGBT multi-junction temperature difference control. It is characterized by including multiple IGBT module units as a control and device under test, a main circuit system for forming a test loop, a small current test system for calculating junction temperature, a water-cooled heat dissipation system containing multiple independent stations, Control system for controlling IGBT grid drive and water-cooling heat dissipation, and data acquisition system for monitoring electrical and thermal parameters.

4 is a flowchart of a power cycle experiment according to the power cycle experiment method provided by the present invention. The implementation steps are as follows: First, set the main circuit current I _c to obtain the collector-emitter voltage V _ce of the IGBT under the set current condition, The conduction power consumption of IGBT is calculated. Next, set the convection heat transfer coefficient, the opening and closing time of the solenoid valve, and calculate the heat dissipation of the water-cooled heat dissipation system. Then, the conduction power consumption of the IGBT and the heat dissipation of the water cooling system are calculated to obtain the total power consumption of the IGBT. According to the theoretical junction temperature calculation formula, the theoretical T _{j of the} IGBT is obtained. If the established experimental requirements are met, the experimental conditions will be used for power cycle experimental debugging. After that, a small current test system is used to collect the saturation voltage drop V _{ce of the} four IGBT modules under test in real time and bring them into the respective K curve calculation formula to calculate the T _{j of} each IGBT module under test. Compare the difference between the experimentally calculated junction temperature and the theoretically calculated junction temperature. If the deviation range is met, the power cycle experiment will be officially started. If the deviation requirements are not met, stop the experiment and continue to adjust the parameters in the experiment until the experimentally calculated junction temperature and the theoretically calculated junction temperature meet the deviation requirements, and the power cycle experiment can be performed.

The specific implementation steps are as follows: First, fix the five modules to the five heat dissipation stations, as shown in Figure 3. Connect the positive electrode of the power supply V1 to one terminal of the resistor R1 according to the circuit diagram 1, connect the other terminal of the resistor to the collector C of IGBT1, connect the collector C of IGBT2 and IGBT3 in parallel to the emitter E of IGBT1, and then connect IGBT2 The emitter E of IGBT3 is connected in parallel to the collector of IGBT4, the emitter E of IGBT4 is connected to the collector of IGBT5, and the emitter E of IGBT5 is connected to the negative electrode of power supply V1.

In the low current test system, the anode of the power supply V2 is connected to the anode of the diode D1, and the cathode of the diode D1 is connected to the emitter E of the IGBT1. Connect the negative pole of power supply V2 to the negative pole of power supply V1.

After that, each parameter is calculated in the order shown in FIG. 5. The set current I _c , the switching time duty cycle D, and the conduction power consumption P _{cond are} calculated according to formula (1). The diameter d of the heat dissipation pipe of the water-cooled heat dissipation system and the flow velocity u of the cooling liquid are taken into formula (5) to calculate the convective heat transfer coefficient α. Then, the heat dissipation Q of the water-cooled plate, the temperature T _{w of the} cooling liquid, and the convection heat transfer coefficient α are taken into equation (6) to calculate the shell temperature T _{c of the} water-cooled plate. After that, the junction temperature T _{j of the} IGBT is calculated according to the conduction power consumption P _cond and the case temperature T _{c of the} water-cooled plate into equation (2).

Carry out the power cycle experiment according to the set switching time, current and other parameters. The timing diagram of each switch is shown in Figure 6. In a power cycle period, when IGBT1 is turned on, the current of IGBT2 and IGBT3 is 0.5I _c , and the current of IGBT4 and IGBT5 is I _c . The water-cooling control solenoid valves of IGBT2 and IGBT5 are turned on, taking away a part of heat generated by IGBT power consumption. The water-cooling control solenoid valves of IGBT3 and IGBT4 are turned off. When IGBT1 is turned off, IGBT2, IGBT3, IGBT4, and IGBT5 are not energized. The water-cooled control solenoid valves of IGBT2, IGBT3, IGBT4, and IGBT5 are all turned on, and quickly cool the four modules.

The low current test system tests the collector-emitter voltage V _{ce of the} four IGBT modules under test during the turn-off of IGBT1. Based on the relationship between collector-emitter voltage V _ce of the module and T _j, respectively, and the IGBT3 IGBT2 V _ce into the formula (2), and the V _ce IGBT5 IGBT4 into the equation (3). It is calculated that the junction temperature difference of IGBT2 in a period is △ T _j2 , the junction temperature difference of IGBT3 is △ T _j3 , the junction temperature difference of IGBT4 is △ T _j4 , and the junction temperature difference of IGBT5 is △ T _j5 . Comparing the experimental calculation T _j with the theoretical calculation T _j , if the difference between the two is large and not within the allowable deviation range, the experimental conditions are reset according to the experimental flow of FIG. 4 to perform the experiment. Until the experimental calculation T _j and the theoretical calculation T _j meet the experimental requirements, the power cycle experiment is started.

Claims

A power cycle experimental device for automotive-grade IGBT multi-junction temperature difference control, characterized in that the device includes: an IGBT module, a main circuit system, a small current test system, a water cooling system, a control system, a data acquisition system; On the water cooling system, connect the main circuit system and the small current test system to the IGBT module; the control system controls the gate of the IGBT1 module and the solenoid valve of the water cooling system; the data acquisition system monitors the current of the main circuit in real time I c , the emitter voltage V ce of the IGBT module to be tested, and the temperature T c of the IGBT module to be tested; and the V ce data under the test current is automatically brought into the K curve to calculate the junction temperature T j of the IGBT module in real time.
An automotive-grade IGBT multi-junction temperature difference control power cycle experiment device according to claim 1, wherein the IGBT module includes five different types of IGBT modules, namely IGBT1, IGBT2, IGBT3, IGBT4 And IGBT5; IGBT1 module with large rated power is used as the switch of the main circuit, and the large rated power is selected to ensure the reliability of the switch; IGBT2, IGBT3, IGBT4, and IGBT5 of the same model and small rated power are used as the device under test to ensure the comparison of the experiment Sex.
The power cycle experiment device for automotive-grade IGBT multi-junction temperature difference control according to claim 1, wherein the main circuit system is composed of a power supply V1, a resistor R1, IGBT1, IGBT2, IGBT3, IGBT4, IGBT5 ; The terminal 1 of the power supply V1 is connected to the 2 terminal of the resistor R1, and the 3 terminal of the resistor R1 is connected to the 4 terminal of the IGBT1; the two output terminals of the IGBT2 and the IGBT3 in parallel, one terminal is connected to the 5 terminal of the IGBT1, and the other terminal is connected to the 10 terminal of the IGBT4 Connection, the 11 end of IGBT4 is connected to the 12 end of IGBT5, and the 13 end of IGBT5 is connected to the 18 end of power supply V1; in this circuit connection method, multiple IGBT modules to be tested use different series and parallel connection circuits to achieve more in the same main circuit IGBT modules have different load currents; the current flowing through IGBT2 and IGBT3 is half of the current in the main circuit, and the current flowing through IGBT4 and IGBT5 is equal to the current in the main circuit;

The calculation formula of the conduction power consumption of the IGBT module is

P cond = D · I c · V ce (I c )

= D · I c · {[V ce_25 ℃ + K V (T j -25 ℃)] + I c · [r ce_25 ℃ + K r (T j -25 ℃)]} (1)

Where D is the duty cycle of the switching time, I c is the load current of the IGBT module, V ce (I c ) is the collector-emitter voltage of the IGBT module, and V ce_25 ° C is the collector-emitter voltage of the IGBT module at 25 ° C , R ce_25 ℃ is the on-resistance of IGBT module at 25 ℃, K V is the temperature coefficient of voltage, K r is the temperature coefficient of resistance;

The calculation formula of the junction temperature T j of the IGBT module is

T j = R th · V ce (I c ) · I c + T c

＝ R th · {[V ce_25 ℃ + K V (T j -25 ℃)] + I c · [r ce_25 ℃ + K r (T j -25 ℃)]} · I c + T c (2)

Among them, R th is the thermal resistance of the IGBT, according to the data manual of the IGBT, the thermal resistance of the module under different experimental conditions is obtained, and T c is the shell temperature of the IGBT;

It can be seen from formula (2) that the junction temperature of the IGBT is related to the power consumption of the IGBT and the shell temperature T c of the water-cooled plate in the water-cooled heat dissipation system.
The power cycle experiment device for automotive-grade IGBT multi-junction temperature difference control according to claim 1, characterized in that the small current test system is mainly composed of a power supply V2 and a diode D1; the positive output terminal 16 of the power supply V2 is The 15 terminal of the diode D1 is connected, and the 14 of the diode D1 is connected to the 5 of the IGBT1; the other output terminal 17 of the power supply is connected to the 13 terminal of the IGBT5.
An automotive-grade IGBT multi-junction temperature difference control power cycle test device according to claim 1, characterized in that: the small current test system provides test current for four IGBT modules under test at the same time; wherein, the test flows into IGBT2 and IGBT3 The current is half of the test current of IGBT4 and IGBT5.
The power cycle experiment device for automotive-grade IGBT multi-junction temperature difference control according to claim 1, characterized in that the small current test system is based on the function of T j and V ce under different small currents in the thermostat experiment Relation curve; get the expression of T j of IGBT at 50mA low current

Expression of T j of IGBT at 100mA low current

Wherein, IGBT2 and the IGBT3 T j T j using test expression under 50mA current calculation; IGBT5 of the IGBT4 and T j T j using the 100mA test current calculation expression.
The power cycle experiment device for automotive-grade IGBT multi-junction temperature difference control according to claim 1, characterized in that: the water-cooled heat dissipation system; can provide a plurality of IGBT modules with cooling liquid for heat dissipation; the cooling liquid is in a circular shape Forced convection heat transfer in the pipeline, the convection heat transfer coefficient ɑ is

Where λ is the thermal conductivity, d is the pipe diameter, u is the liquid flow rate, ρ is the liquid density, μ is the liquid viscosity, c p is the specific heat capacity, n is 0.4 when the fluid is heated, and 0.3 for cooling;

According to the convection heat transfer coefficient formula, when other parameters are fixed, α is proportional to the 0.8 power of u, indicating that increasing the flow velocity u is beneficial to increase α;

When other parameters are fixed, α is inversely proportional to the 0.2 power of d, indicating that reducing the pipe diameter d is beneficial to increase α;

The formula of heat dissipation Q of water cooling system is

Q ＝ a · A · (T c -T w ) (6)

Among them, Q is the heat dissipation of the water-cooled plate, unit W; a is the convection heat transfer coefficient, unit W / (m 2. ℃); A is the heat conduction area, unit m 2 ; T c is the temperature of the water-cooled plate, that is, the shell temperature of the IGBT , Unit ℃; T w is the coolant temperature, unit ℃;

According to the heat dissipation formula of the water-cooled heat dissipation system, when other parameters are fixed, Q is proportional to a; therefore, the convection heat transfer coefficient a is adjusted by changing the pipe diameter d of the water-cooled plate and the liquid flow rate u in the water-cooled heat dissipation system , And then control the heat dissipation of the water-cooled heat dissipation system, thereby controlling the IGBT case temperature T c .
The power cycle experiment device for automotive-grade IGBT multi-junction temperature difference control according to claim 1, characterized in that the water-cooled heat dissipation system includes five water-cooled stations; each station is equipped with an independent solenoid valve According to the heat dissipation requirements of the IGBT module, independently control the on and off of each water-cooling station solenoid valve, and then control the on and off time of the cooling liquid.
The power cycle experiment device for automotive-grade IGBT multi-junction temperature difference control according to claim 1, characterized in that: the control system is controlled by a single-chip microcomputer, which can control the switching module IGBT1 and the water-cooling cooling system solenoid valve On and off.
The power cycle experiment device for automotive-grade IGBT multi-junction temperature difference control according to claim 1, wherein the data acquisition system can simultaneously monitor the current I c of the main circuit and the emitter of the IGBT module to be tested Data such as voltage V ce and the temperature of the IGBT module case T c to be tested, and store the collected data in real time in the computer;

The data collection system automatically extracts the data of the emitter voltage V ce of the IGBT module to be tested under low current, and calculates the T j change of the IGBT module according to the K curve of the IGBT module.