Screening method of hybrid semiconductor device
By screening the drift amount and drift rate of the switch tube in hybrid semiconductor devices and connecting in parallel, the problem of poor current sharing effect and power loss caused by characteristic parameter drift in the device is solved, and the effect of low power loss and high switching tube utilization is achieved.
Patent Information
- Application Number
- CN202510322008.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-18
- Publication Date
- 2025-06-20
AI Technical Summary
In hybrid semiconductor devices, the characteristic parameters drift between the same type of semiconductor devices, resulting in poor current sharing effect and power loss.
By measuring the drift and drift rate of the switch tube, filter out the drift rate of the switch tube within the specified range and connect them in parallel to control the drift degree of threshold voltage, on-resistance and saturated on-voltage drop.
The switching delay between different switch tubes is achieved to meet the requirements, the shunt effect is good, the overall power loss is low, and the switching tube utilization rate is high.
Smart Images
Figure CN120184031A_ABST
Abstract
Description
Technical Field
[0001] This case involves a hybrid semiconductor device, especially a screening method for hybrid semiconductor devices. Background Art
[0002] Semiconductor devices are widely used in new energy vehicles, wind power generation, power supplies and other places. In high-power application scenarios, a single semiconductor device often cannot meet the demand for large current. Therefore, multiple semiconductor devices are connected in parallel to shunt the large current, so that multiple semiconductor devices share the power loss together, which can effectively improve the efficiency, increase the current capacity, and reduce the conduction loss at the same time.
[0003] In current technologies, in order to save costs and reduce power loss at the same time, different types of semiconductor devices are mixed and connected in parallel, such as insulated gate bipolar transistors and wide bandgap semiconductor devices connected in parallel. In a hybrid semiconductor device, the characteristic parameters (such as threshold voltage, saturation voltage drop, and on-resistance, etc.) between the same type of semiconductor devices need to be consistent to achieve a good current sharing effect, while between different types of semiconductor devices, fixed switching timings and current sharing ratios are required to achieve better power loss. Among them, the switching timing is related to the threshold voltage, and the current sharing ratio is related to the saturation voltage drop and on-resistance. However, due to factors such as process difficulty and technical limitations, the characteristic parameters between the hybrid devices connected in parallel, such as threshold voltage, saturation voltage drop, and on-resistance, have a certain degree of drift, resulting in differences between the current sharing effect between the same type of semiconductor devices, the switching timings and current sharing ratios between different types of semiconductor devices and the expected values, which cannot meet the matching requirements, resulting in the problems of uneven current sharing and large power loss in hybrid semiconductor devices.
[0004] Therefore, it is necessary to develop a screening method for hybrid semiconductor devices to solve the problems faced by the prior art. Summary of the Invention
[0005] The purpose of this case is to provide a screening method for hybrid semiconductor devices, which measures whether the drift amount of the switching tube is within the corresponding drift amount range and measures whether the drift rate is within the corresponding drift rate range, so as to connect in parallel the switching tubes with the drift amount within the drift amount range and / or the switching tubes with the drift rate within the drift rate range. Therefore, the drift degrees of the threshold voltage, on-resistance and saturation on-state voltage drop of the switching tubes connected in parallel are similar, so the turn-on and turn-off delays between different switching tubes can meet the requirements, and the current sharing effect is also relatively good, making the overall power loss of the hybrid semiconductor device lower and the utilization rate of the switching tubes higher.
[0006] To achieve the above object, the present case provides a screening method for screening hybrid semiconductor devices, where the hybrid semiconductor devices include a plurality of first switching transistors and a plurality of second switching transistors, and the screening method includes the following steps. First, measure the first drift amount and the first drift rate of each first switching transistor, and measure the second drift amount and the second drift rate of each second switching transistor. Then, confirm whether the first drift rate of each first switching transistor and the second drift rate of each second switching transistor are within the drift rate range, and screen out the corresponding first switching transistors and the corresponding second switching transistors whose first drift rate and second drift rate are within the drift rate range. Then, confirm whether the first drift amount of the remaining first switching transistors and the second drift amount of the remaining second switching transistors are within the drift amount range, and screen out the corresponding first switching transistors and the corresponding second switching transistors whose first drift amount and second drift amount are within the drift amount range. Then, connect in parallel the first switching transistors and the second switching transistors corresponding to the first drift rate and the second drift rate within the drift rate range, and / or the first switching transistors and the second switching transistors corresponding to the first drift amount and the second drift amount within the drift amount range.
[0007] According to one embodiment of the present invention, the screening method further includes measuring the first threshold voltage of each first switching transistor and measuring the second threshold voltage of each second switching transistor; where the first drift amount is the first threshold voltage minus the first typical voltage, and the second drift amount is the second threshold voltage minus the second typical voltage.
[0008] According to one embodiment of the present invention, the screening method further includes measuring the saturation conduction voltage drop of each first switching transistor and measuring the on-resistance of each second switching transistor; where the first drift rate is the difference between the saturation conduction voltage drop and the typical saturation conduction voltage drop, divided by the difference between the highest saturation conduction voltage drop and the lowest saturation conduction voltage drop; the second drift rate is the difference between the on-resistance and the typical on-resistance, divided by the difference between the highest on-resistance and the lowest on-resistance.
[0009] According to one embodiment of the present invention, the screening method further includes setting screening grid coordinates, with the drift amount as the X-axis of the screening grid coordinates and the drift rate as the Y-axis of the screening grid coordinates; marking the first drift rate and the first drift amount corresponding to each first switching transistor into the screening grid coordinates; marking the second drift rate and the second drift amount corresponding to each second switching transistor into the screening grid coordinates.
[0010] According to one embodiment of the present invention, the screening method further includes that the screening grid coordinates include n first screening precisions, m second screening precisions, a first screening step size, and a second screening step size, where the first screening step size is composed of i first screening precisions, the second screening step size is composed of j second screening precisions, where i = 1, 2... n, j = 1, 2... m.
[0011] According to one embodiment of the present invention, the screening method further includes that n is the smallest positive integer satisfying the first inequality, where the first inequality is n×△x / 2≥max{|△Vth|}, where △x is the first screening accuracy, and max{|△Vth|} is the maximum value of the absolute values of all first drift amounts and all second drift amounts; m is the smallest positive integer satisfying the second inequality, where the second inequality is m×△y / 2≥max{|△Vds|}, where △y is the second screening accuracy, and max{|△Vds|} is the maximum value of the absolute values of all first drift rates and all second drift rates.
[0012] According to one embodiment of the present invention, the screening method further includes that in the grid corresponding to any first screening accuracy in the first screening step and the corresponding second screening accuracy in the second screening step, it is confirmed whether there is at least one first switching tube and at least one second switching tube in the grid; then, for the first switching tubes and second switching tubes not located in the grid in the previous step, continue to screen, increase the number of the second screening accuracies of the grid to increase the second screening step of the grid, and confirm again whether there is at least one first switching tube and at least one second switching tube in the grid after increasing the second screening step; then, the number of the second screening accuracies increases from 1 to m, and for each increase of 1 in the number of the second screening accuracies, the previous step is re-executed to obtain the grid with the maximum second screening step.
[0013] According to one embodiment of the present invention, the screening method further includes increasing the number of the first screening accuracies of the grid to increase the first screening step of the grid, and repeating the steps: in the grid corresponding to any first screening accuracy in the first screening step and the corresponding second screening accuracy in the second screening step, it is confirmed whether there is at least one first switching tube and at least one second switching tube in the grid; and the step: the number of the second screening accuracies increases from 1 to m, and for each increase of 1 in the number of the second screening accuracies, the previous step is re-executed to obtain the grid with the maximum second screening step; then, the number of the first screening accuracies increases from 1 to n, and for each increase of 1 in the number of the first screening accuracies, the step of increasing the number of the first screening accuracies of the grid to increase the first screening step of the grid is executed once to obtain the grid with the maximum first screening step; then, determine the grid corresponding to any second screening step and / or any first screening step, and select the corresponding first switching tube and second switching tube in the grid to be connected in parallel.
[0014] According to one embodiment of the present invention, the screening method further includes that the first screening accuracy is a real number between 0 and the maximum value of the absolute values of all first drift rates and all second drift rates, and the second screening accuracy is a real number between 0 and 1; wherein the number of screening times of the screening method is inversely proportional to the first screening accuracy and the second screening accuracy.
[0015] According to one embodiment of the present invention, the screening method further includes that a plurality of first switching tubes are insulated gate bipolar transistors, and a plurality of second switching tubes are wide bandgap semiconductor devices. Description of the Drawings
[0016] Figure 1 It is a circuit architecture diagram of the hybrid semiconductor device of this case applied to an inverter;
[0017] Figure 2A It is Figure 1 A schematic diagram of a measurement circuit of the first threshold voltage of the first switching tube in the hybrid semiconductor device shown;
[0018] Figure 2B It is Figure 1 A schematic diagram of a measurement circuit of the second threshold voltage of the second switching tube in the hybrid semiconductor device shown;
[0019] Figure 2C It is Figure 1 A schematic diagram of a measurement circuit of the saturation conduction voltage drop of the first switching tube in the hybrid semiconductor device shown;
[0020] Figure 2D It is Figure 1 A schematic diagram of a measurement circuit of the on-resistance of the second switching tube in the hybrid semiconductor device shown;
[0021] Figures 3A to 3H It is applied to Figure 1 A schematic diagram of the screening grid coordinates of the hybrid semiconductor device shown; and
[0022] Figure 4 It is Figure 1 A flowchart of the screening method of the hybrid semiconductor device shown.
[0023] Among them, the description of the reference numerals is as follows:
[0024] 1: Hybrid semiconductor device
[0025] 2: Inverter
[0026] 3: First switching tube
[0027] 4: Second switching tube
[0028] 5: Controller
[0029] 61: First grid
[0030] 62: Second grid
[0031] 63: Third grid
[0032] 64: Fourth grid
[0033] 65: The fifth grid
[0034] 66: The sixth grid
[0035] 67: The seventh grid
[0036] 68: The eighth grid
[0037] 69: The ninth grid
[0038] 610: The tenth grid
[0039] S1 - S4: Steps Detailed implementation manners
[0040] Some typical embodiments reflecting the features and advantages of this case will be described in detail in the following description. It should be understood that this case can have various variations in different aspects, all of which do not deviate from the scope of this case, and the descriptions and drawings therein are essentially for illustrative purposes and not for limiting this case.
[0041] Please refer to Figure 1 , which is a circuit architecture diagram of the hybrid semiconductor device of this case applied to an inverter. As shown in the figure, the hybrid semiconductor device 1 of this case is applied to the inverter 2, and the inverter 2 may include multiple hybrid semiconductor devices 1 and a controller 5, such as Figure 1 the six hybrid semiconductor devices 1 shown in. In this embodiment, the hybrid semiconductor device 1 includes two first switching tubes 3 and one second switching tube 4, where the switching tube type of the first switching tube 3 is different from that of the second switching tube 4. For example, the first switching tube 3 can be an Insulated - Gate Bipolar Transistor (IGBT), and the second switching tube 4 can be a wide - bandgap semiconductor device, which is a Metal - Oxide - Semiconductor Field - Effect Transistor (MOSFET) composed of a wide - bandgap semiconductor material. The wide - bandgap semiconductor materials mainly include silicon carbide (SiC), gallium nitride (GaN), zinc oxide (ZnO), diamond, aluminum nitride (AlN), etc.
[0042] The screening method of this case is used to screen out matching switching tubes for parallel connection to form Figure 1 the first switching tubes 3 and the second switching tubes 4 connected in parallel as shown in, and the number of the first switching tubes 3 and the second switching tubes 4 can be but is not limited to Figure 1The two first switching transistors 3 and one second switching transistor 4 shown may also be different numbers of first switching transistors 3 and second switching transistors 4. The following will further describe the screening method by which the inverter 2 uses the controller 5 to screen the switching transistors of the hybrid semiconductor device 1.
[0043] The controller 5 is used to measure the first drift amount and the first drift rate of each first switching transistor 3, and measure the second drift amount and the second drift rate of each second switching transistor 4. In this embodiment, the controller 5 measures the first threshold voltage of the first switching transistor 3, and subtracts the first typical voltage from the first threshold voltage to obtain the first drift amount of the first switching transistor 3; the controller 5 measures the second threshold voltage of the second switching transistor 4, and subtracts the second typical voltage from the second threshold voltage to obtain the second drift amount of the second switching transistor 4, where the first typical voltage and the second typical voltage are respectively the most representative or most common voltage values or the expected voltage values of the first switching transistor 3 and the second switching transistor 4 under standard or normal operating conditions, to reflect the voltage concentration levels of the first switching transistor 3 and the second switching transistor 4. For example, they may be the voltage averages of multiple first switching transistors 3 and multiple second switching transistors 4.
[0044] In this embodiment, the controller 5 measures the saturation conduction voltage drop of each first switching transistor 3, and divides the difference between the saturation conduction voltage drop and the typical saturation conduction voltage drop by the difference between the highest saturation conduction voltage drop and the lowest saturation conduction voltage drop to obtain the first drift rate of the first switching transistor 3, where the typical saturation conduction voltage drop is the most representative or most common or expected saturation conduction voltage drop of the first switching transistor 3 under standard or normal operating conditions, to reflect the saturation conduction voltage drop concentration level of the first switching transistor 3. For example, it may be the average saturation conduction voltage drop of multiple first switching transistors 3. The controller 5 further measures the on-resistance of each second switching transistor 4, and divides the difference between the on-resistance and the typical on-resistance by the difference between the highest on-resistance and the lowest on-resistance to obtain the second drift rate, where the typical on-resistance is the most representative or most common or expected on-resistance of the second switching transistor 4 under standard or normal operating conditions, to reflect the on-resistance concentration level of the second switching transistor 4. For example, it may be the average on-resistance of multiple second switching transistors 4. In this embodiment, the controller 5 can obtain the typical values of the various parameters of specific components by referring to the data sheet. The highest saturation conduction voltage drop of the first switching transistor 3 is the maximum value among all the saturation conduction voltage drop values, the lowest saturation conduction voltage drop of the first switching transistor 3 is the minimum value among all the saturation conduction voltage drop values, the highest on-resistance is the maximum value among all the on-resistance values, and the lowest on-resistance is the minimum value among all the on-resistance values.
[0045] In this embodiment, the controller 5 includes a first measurement circuit 51 and a second measurement circuit 52. The first measurement circuit 51 is used to measure the first threshold voltage of the first switching transistor 3 and the second threshold voltage of the second switching transistor 4. For example, Figure 2A and Figure 2B as shown, the first measurement circuit 51 can be connected to the first switching transistor 3 to measure the first threshold voltage of the first switching transistor 3, and can be connected to the second switching transistor 4 to measure the second threshold voltage of the second switching transistor 4. In this embodiment, the first measurement circuit 51 gradually increases the power supply voltage (such as Figure 2A Vge shown and Figure 2B Vgs shown), and uses an ammeter (such as Figure 2A shown and Figure 2B ammeter A shown) to monitor the drain current in real time, and takes the power supply voltage value corresponding to the start of the linear rise of the drain current as the first threshold voltage and the second threshold voltage. The second measurement circuit 52 is used to measure the saturation conduction voltage drop of the first switching transistor 3 and the on-resistance of the second switching transistor 4. For example, Figure 2C and Figure 2D as shown, the second measurement circuit 52 can be connected to the first switching transistor 3 to measure the saturation conduction voltage drop of the first switching transistor 3. In this embodiment, the second measurement circuit 52 measures the conduction voltage of the first switching transistor 3 at a given collector saturation current during conduction as the saturation conduction voltage drop, and can be connected to the second switching transistor 4 to measure the on-resistance of the second switching transistor 4. The ratio of the drain-source voltage to the drain current at a given drain current during conduction is measured and calculated as the on-resistance.
[0046] As shown in Table 1, it is the actual measurement data of the first threshold voltage, saturation conduction voltage drop of the first switching transistor 3, the second threshold voltage of the second switching transistor 4, and the on-resistance. In Table 1, a total of twenty first switching transistors 3 and ten second switching transistors 4 are measured. In this embodiment, the controller 5 can determine that the first typical voltage of the first switching transistor 3 is 6.4, the second typical voltage of the second switching transistor 3 is 2.5, the typical saturation conduction voltage drop of the first switching transistor 3 is 1.55, and the typical on-resistance of the second switching transistor 3 is 8.8 by referring to the data manuals of the first switching transistor 3 and the second switching transistor 4. As shown in Table 1, the highest saturation conduction voltage drop is 1.8, and the lowest saturation conduction voltage drop is 1.2. As shown in Table 1, the highest on-resistance is 9.2, and the lowest on-resistance is 8.48.
[0047] Table 1 Actual measurement data of the first threshold voltage, saturation conduction voltage drop of the first switching transistor, the second threshold voltage of the second switching transistor, and the on-resistance
[0048]
[0049] As shown in Table 2, it is the actual measurement data corresponding to the first drift amount, the first drift rate of the first switching transistor 3, the second drift amount of the second switching transistor 4, and the second drift rate in Table 1.
[0050] Table 2 shows the actual measurement data corresponding to the first drift amount, the first drift rate of the first switching transistor, the second drift amount of the second switching transistor, and the second drift rate in Table 1
[0051]
[0052] A screening grid coordinate is further set in the controller 5, where Figures 3A to 3H is for application to Figure 1 a schematic diagram of the screening grid coordinate of the hybrid semiconductor device shown. In Figure 3A , the X-axis of the screening grid coordinate is the drift amount (including the first drift amount of the first switching transistor 3 and the second drift amount of the second switching transistor 4), and the Y-axis of the screening grid coordinate is the drift rate (including the first drift rate of the first switching transistor 3 and the second drift rate of the second switching transistor 4). And the screening grid coordinate includes n first screening precisions, m second screening precisions, a first screening step size, and a second screening step size. Each first screening precision constitutes each grid coordinate in the X-axis of the screening grid coordinate, where the first screening precision is a real number between 0 and the maximum value of the absolute values of all the first drift amounts and all the second drift amounts. When the value of the first screening precision is smaller, the screening precision is higher and the number of screening times is more. For example, in Table 2, the maximum value of the absolute values of all the first drift amounts and all the second drift amounts is 0.71. So the value of the first screening precision is any value greater than 0 and less than or equal to 0.71. For example, the first screening precision in this embodiment can be, for example, Figure 3A 0.2 in
[0053] n×△x / 2≥max{|△Vth|}…(1)
[0054] Among them, △x is the first screening accuracy, △Vth is the sum of all first drift amounts and all second drift amounts, max{|△Vth|} is the maximum absolute value among all first drift amounts and all second drift amounts of the measured switching transistor. It can also be understood that n is the smallest positive integer that satisfies the condition that the first screening accuracy multiplied by n / 2 is greater than or equal to the maximum absolute value of all first drift amounts and all second drift amounts. Therefore, it can be known that when the maximum absolute value of all first drift amounts and all second drift amounts is 0.71 and the first screening accuracy is 0.2, then n is 8. Each second screening accuracy constitutes each grid coordinate on the Y-axis of the screening grid coordinates. Among them, the second screening accuracy is a real number between 0 and 1. When the value of the second screening accuracy is smaller, the screening accuracy is higher, but the number of screenings is more. In this embodiment, the second screening accuracy can be, for example, Figure 3A 0.2 in Figure 3A . The second screening step size is composed of j second screening accuracies, where j = 1, 2... m. In this embodiment, m is the smallest positive integer that satisfies the following second inequality (2).
[0055] m×△y / 2≥max{|△V ds |}…(2)
[0056] Among them, △y is the second screening accuracy, △Vds is the sum of all first drift rates and all second drift rates, max{|△Vds|} is the maximum absolute value among all first drift rates and all second drift rates of the measured switching transistor. It can also be understood that m is the smallest positive integer that satisfies the condition that the second screening accuracy multiplied by m / 2 is greater than or equal to the maximum absolute value of all first drift rates and all second drift rates. Therefore, it can be known that when the maximum absolute value of all first drift amounts and all second drift amounts is 0.58 and the second screening accuracy is 0.2, then m is 6. As can be seen from the above, in this embodiment, the first screening step size of the screening grid coordinates is i first screening accuracies, where i = 1, 2... 8; the second screening step size is j second screening accuracies, where j = 1, 2... 6, and the screening grid coordinates can be formed Figure 3A thereby.
[0057] The controller 5 marks the first drift rate and the first drift amount of each first switching transistor 3 it measures into Figure 3A the screening grid coordinates, and marks the second drift rate and the second drift amount of each second switching transistor 4 it measures into Figure 3A the screening grid coordinates, where Figure 3A each point in Figure 3A represents the drift rate and the drift amount corresponding to each switching transistor, and the number next to each point represents its corresponding number.
[0058] The controller 5 further confirms whether the first drift rate of each first switching tube 3 marked into the screening grid coordinates and the second drift rate of each second switching tube 4 marked into the screening grid coordinates are within the drift rate range, and screens out the first switching tubes and the corresponding second switching tubes corresponding to the first drift rate and the second drift rate within the drift rate range. In this embodiment, the controller 5 further performs screening by the following steps. First, in the grid corresponding to any first screening accuracy in the first screening step length and the corresponding second screening accuracy in the second screening step length, and confirm whether there are two first switching tubes 3 and one second switching tube 4 in the grid at the same time. For example, when initializing i and j, that is, when i = 1 and j = 1, the first screening step length is 0.2 and the second screening step length is 0.2. For example Figure 3A , in the grid with the X-axis from 0 to 0.2 and the Y-axis from 0 to 0.2, there is a second switching tube 4 numbered 1, a first switching tube 3 numbered 3, and a first switching tube 3 numbered 5. There are two first switching tubes 3 and one second switching tube 4 in this grid, which is hereinafter referred to as the first grid 61. Therefore, the two first switching tubes 3 and one second switching tube 4 in the first grid 61 can be screened out for parallel connection; for example, when i = 1 and j = 1, in the grid with the X-axis from 0.2 to 0.4 and the Y-axis from -0.2 to 0, there is a second switching tube 4 numbered 2, a first switching tube 3 numbered 6, and a first switching tube 3 numbered 17. There are two first switching tubes 3 and one second switching tube 4 in this grid, which is hereinafter referred to as the second grid 62. Therefore, the two first switching tubes 3 and one second switching tube 4 in the second grid 62 can be screened out for parallel connection. The second grid 62 also has a second switching tube 4 numbered 7. In the second grid 62, the second switching tube 4 numbered 2 or the second switching tube 4 numbered 7 can be selected. In this embodiment, the second switching tube 4 numbered 2 is selected in the second grid 62, and the second switching tube 4 numbered 7 continues to be screened by the following steps. Then, continue to screen the first switching tubes 3 and the second switching tubes 4 that are not in the previous steps, where the number of the second screening accuracies of the grid in the previous steps is increased to increase the second screening step length of the grid. For example, when i = 1 and j = 2, the number of the second screening accuracies is 2, the first screening step length is 0.2, and the second screening step length is 0.4, as Figure 3B shown, and confirm again whether there are two first switching tubes 3 and one second switching tube 4 in the grid after increasing the second screening step length. In Figure 3B , no grid with two first switching tubes 3 and one second switching tube 4 at the same time is screened out.
[0059] Increase the number of the second screening accuracies of the grid in the previous steps to increase the second screening step length of the grid, as Figure 3CAs shown, for example, when i = 1 and j = 3, the number of the second screening precisions is 3, the first screening step size is 0.2 and the second screening step size is 0.6. In the grid with the X-axis ranging from -0.2 to 0 and the Y-axis ranging from -0.3 to 0.3, there are the second switching tube 4 numbered 3, the first switching tube 3 numbered 4, and the first switching tube 3 numbered 1. There are two first switching tubes 3 and one second switching tube 4 in this grid, hereinafter referred to as the third grid 63. Therefore, the two first switching tubes 3 and one second switching tube 4 in the third grid 63 can be screened out for parallel connection. The third grid 63 also has the first switching tube 3 numbered 9 and the first switching tube 3 numbered 12. Two of the first switching tubes 3 numbered 1, 4, 9, and 12 in the third grid 63 can be selected for parallel connection. In this embodiment, the first switching tubes 3 numbered 1 and 4 in the third grid 63 are randomly selected, while the first switching tubes 3 numbered 9 and 12 continue to be screened through the following steps. In this embodiment, as Figure 3C shown in the grid form, in the screening grid, the coordinate axis X = 0 or the coordinate axis Y = 0 represents that the drift amount is 0 or the drift rate is 0, which means that the measured parameters (the first threshold voltage, the second threshold voltage, the saturation conduction voltage drop, the conduction resistance) of the first switching tube 3 and / or the second switching tube 4 are consistent with the typical values. When using Figure 3C the screening method to construct a coordinate system by diverging from the origin "0" of the coordinate axis to both positive and negative semi-axes, it can also be understood that a coordinate system is constructed by diverging from the drift amount of 0 or the drift rate of 0 to both positive and negative semi-axes, so as to make the drift amount or drift rate of the screened switching tubes as close to 0 as possible, so that the measured parameters (the first threshold voltage, the second threshold voltage, the saturation conduction voltage drop, the conduction resistance) of the first switching tube 3 and / or the second switching tube 4 can be as close to the typical values as possible.
[0060] Increase the number of the second screening precisions of the grid in the foregoing steps to increase the second screening step size of the grid, as Figure 3DAs shown, for example, when i = 1 and j = 4, the number of the second screening precisions is 4, the first screening step size is 0.2 and the second screening step size is 0.8. In the grid where the X-axis ranges from -0.2 to 0 and the Y-axis ranges from -0.4 to 0.4, there are the second switching tube 4 numbered 4, the first switching tube 3 numbered 9 and the first switching tube 3 numbered 12. There are two first switching tubes 3 and one second switching tube 4 in this grid, hereinafter referred to as the fourth grid 64. Therefore, the two first switching tubes 3 and one second switching tube 4 in the fourth grid 64 can be screened out for parallel connection. And in the grid where the X-axis ranges from 0 to 0.2 and the Y-axis ranges from -0.4 to 0.4, there are the second switching tube 4 numbered 5, the first switching tube 3 numbered 2 and the first switching tube 3 numbered 14. There are two first switching tubes 3 and one second switching tube 4 in this grid, hereinafter referred to as the fifth grid 65. Therefore, the two first switching tubes 3 and one second switching tube 4 in the fifth grid 65 can be screened out for parallel connection. And in the grid where the X-axis ranges from 0.6 to 0.8 and the Y-axis ranges from -0.4 to 0.4, there are the second switching tube 4 numbered 10, the first switching tube 3 numbered 10 and the first switching tube 3 numbered 15. There are two first switching tubes 3 and one second switching tube 4 in this grid, hereinafter referred to as the sixth grid 66. Therefore, the two first switching tubes 3 and one second switching tube 4 in the sixth grid 66 can be screened out for parallel connection.
[0061] Next, repeat the above steps for screening. For example, when i = 1 and j = 5, the first screening step size is 0.2 and the second screening step size is 1, as Figure 3E shown, in the grid where the X-axis ranges from -0.6 to -0.4 and the Y-axis ranges from -0.5 to 0.5, there are the second switching tube 4 numbered 9, the first switching tube 3 numbered 7 and the first switching tube 3 numbered 13. There are two first switching tubes 3 and one second switching tube 4 in this grid, hereinafter referred to as the seventh grid 67. Therefore, the two first switching tubes 3 and one second switching tube 4 in the seventh grid 67 can be screened out for parallel connection. Until the number of the increased second screening precisions makes the second screening step size reach the maximum value. When i = 1 and j = 6, the first screening step size is 0.2 and the second screening step size is 1.2, as Figure 3FAs shown, in the grid where the X-axis ranges from -0.4 to -0.2 and the Y-axis ranges from -0.6 to 0.6, there are a second switching tube 4 numbered 6, a first switching tube 3 numbered 8, and a first switching tube 3 numbered 20. There are two first switching tubes 3 and one second switching tube 4 in this grid, hereinafter referred to as the eighth grid 68. Therefore, the two first switching tubes 3 and one second switching tube 4 in the eighth grid 68 can be selected for parallel connection. The eighth grid 68 also has a first switching tube 3 numbered 16. Two of the first switching tubes 3 numbered 8, 16, and 20 can be selected for parallel connection. In this embodiment, the first switching tubes 3 numbered 8 and 20 are selected for the eighth grid 68, and the first switching tube 3 numbered 16 continues to be screened.
[0062] When the second screening step size reaches the maximum value, the controller 5 further confirms whether the first drift amount of the remaining first switching tubes 3 marked in the screening grid coordinates and the second drift amount of each second switching tube 4 marked in the screening grid coordinates are within the drift amount range, and screens out the first switching tubes and the corresponding second switching tubes corresponding to the first drift amount and the second drift amount within the drift amount range. In this embodiment, the controller 5 further performs screening using the following steps. First, increase the number of the first screening precisions of the grid to increase the first screening step size of the grid, and repeatedly increase the number of the second screening precisions to increase the second screening step size of the grid for screening. For example, screen the grid with i = 2 and j = 1, then screen the grid with i = 2 and j = 2, the grid with i = 2 and j = 3, until the grid with i = 2 and j = 6 is screened. Next, screen the grid with i = 3 and j = 1, then screen the grid with i = 3 and j = 2, the grid with i = 3 and j = 3, until the grid with i = 3 and j = 6 is screened. Finally, repeatedly increase the number of the first screening precisions of the grid for screening to obtain the grid with the maximum first screening step size. For example, finally screen to the grid with i = 8 and j = 6, and the number of screening times is 8 * 6 = 48 times, and select the corresponding first switching tube 3 and second switching tube 4 in this grid for parallel connection. For example, in this embodiment, when i = 4 and j = 5, the first screening step size is 0.8 and the second screening step size is 1, as Figure 3G As shown, in the grid where the X-axis ranges from -0.4 to 0.4 and the Y-axis ranges from -0.5 to 0.5, there are a second switching tube 4 numbered 7, a first switching tube 3 numbered 19, and a first switching tube 3 numbered 11. There are two first switching tubes 3 and one second switching tube 4 in this grid, hereinafter referred to as the ninth grid 69. Therefore, the two first switching tubes 3 and one second switching tube 4 in the ninth grid 69 can be selected for parallel connection. When i = 7 and j = 2, the first screening step size is 1.4 and the second screening step size is 0.4, as Figure 3HAs shown, in a grid where the X-axis ranges from -0.7 to 0.7 and the Y-axis ranges from -0.6 to -0.2, there are a second switching transistor 4 numbered 8, a first switching transistor 3 numbered 18, and a first switching transistor 3 numbered 16. There are two first switching transistors 3 and one second switching transistor 4 in this grid, hereinafter referred to as the tenth grid 610. Therefore, the two first switching transistors 3 and one second switching transistor 4 in the tenth grid 610 can be selected for parallel connection. In some embodiments, in order to reduce the computational burden of the controller 5, the screening can be stopped after enough switch transistors that can be connected in parallel are screened out.
[0063] As shown in Table 3, it is a list of screening combinations where the ratio of the number of first switching transistors 3 to second switching transistors 4 simultaneously present in the grid is 2:1.
[0064] Table 3 List of screening combinations where the ratio of the number of first switching transistors to second switching transistors simultaneously present in the grid is 2:1
[0065]
[0066] Of course, the ratio of the number of first switching transistors 3 to second switching transistors 4 simultaneously present in the grid is not limited to the above two first switching transistors 3 and one second switching transistor 4. In one embodiment, the ratio of the number of first switching transistors 3 to second switching transistors 4 simultaneously present in the grid can be one first switching transistor 3 and one second switching transistor 4. The screening method is similar to the above screening method, so it will not be elaborated here.
[0067] As shown in Table 4, it is a list of screening combinations where the ratio of the number of first switching transistors 3 to second switching transistors 4 simultaneously present in the grid is 1:1.
[0068] Table 4 List of screening combinations where the ratio of the number of first switching transistors to second switching transistors simultaneously present in the grid is 1:1
[0069]
[0070] The smaller the screening order of the selected switch transistor combination, the higher the screening accuracy, the closer the parameters between the switch transistors, and the better the parallel connection effect. In other words, the number of screening times of the screening method is inversely proportional to the first screening accuracy and the second screening accuracy.
[0071] In this embodiment, the controller 5 connects in parallel the first switching transistors 3 and second switching transistors 4 corresponding to the first drift rate and the second drift rate within the drift rate range, and / or the first switching transistors 3 and second switching transistors 4 with the first drift amount and the second drift amount within the drift amount range.
[0072] Please refer to Figure 4 which is Figure 1Flowchart of the screening method for the hybrid semiconductor device shown. First, step S1 is executed to measure the first drift amount and the first drift rate of each first switching transistor 3, and measure the second drift amount and the second drift rate of each second switching transistor 4. Then, step S2 is executed to confirm whether the first drift rate of each first switching transistor 3 and the second drift rate of each second switching transistor 4 are within the drift rate range, and screen out the first switching transistor 3 and the corresponding second switching transistor 4 whose first drift rate and second drift rate are within the drift rate range. Then, step S3 is executed to confirm whether the first drift amount of the remaining first switching transistors 3 and the second drift amount of the remaining second switching transistors 4 are within the drift amount range, and screen out the first switching transistor 3 and the corresponding second switching transistor 4 whose first drift amount and second drift amount are within the drift amount range. Then, step S4 is executed to connect in parallel the first switching transistor 3 and the second switching transistor 4 corresponding to the first drift rate and the second drift rate within the drift rate range, and / or the first switching transistor 3 and the second switching transistor 4 whose first drift amount and second drift amount are within the drift amount range.
[0073] In summary, the screening method for the hybrid semiconductor device in this case uses measuring whether the drift amount of the switching transistor is within the corresponding drift amount range and measuring whether the drift rate is within the corresponding drift rate range to connect in parallel the switching transistors whose drift amount is within the drift amount range and / or the switching transistors whose drift rate is within the drift rate range. Therefore, the drift degrees of the threshold voltage, on-resistance, and saturation on-voltage drop of the switching transistors connected in parallel are similar. Thus, the turn-on and turn-off delays between different switching transistors can meet the requirements, and the current sharing effect is also relatively good, resulting in a lower power loss of the overall hybrid semiconductor device, a higher utilization rate of the switching transistors, and the screening step size can be set according to requirements, thereby adjusting the screening accuracy and screening speed.
Claims
1. A screening method, characterized in that: Used to screen a hybrid semiconductor device, wherein the hybrid semiconductor device includes a plurality of first switch tubes and a plurality of second switch tubes, wherein the screening method includes: S1: measuring a first drift amount and a first drift rate of each of the first switch tubes, and measuring a second drift amount and a second drift rate of each of the second switch tubes; S2: confirming whether the first drift rate of each of the first switch tubes and the second drift rate of each of the second switch tubes are within a drift rate range, and selecting the first switch tubes and the second switch tubes corresponding to the first drift rate and the second drift rate within the drift rate range; S3: confirming whether the first drift of the remaining first switch tubes and the second drift of the remaining second switch tubes are within a drift range, and selecting the first switch tubes and the second switch tubes corresponding to the first drift and the second drift within the drift range; and S4: connecting in parallel the first switch tube and the second switch tube corresponding to the first drift rate and the second drift rate being within the drift rate range, and / or the first switch tube and the second switch tube corresponding to the first drift amount and the second drift amount being within the drift amount range.
2. The screening method according to claim 1, characterized in that The screening method further includes: measuring a first threshold voltage of each of the first switch tubes, and measuring a second threshold voltage of each of the second switch tubes; wherein the first drift amount is the first threshold voltage minus a first typical voltage, and the second drift amount is the second threshold voltage minus a second typical voltage.
3. The screening method according to claim 1, characterized in that The screening method further includes: measuring a saturation on-state voltage drop of each of the first switching tubes, and measuring an on-state resistance of each of the second switching tubes; wherein the first drift rate is the difference between the saturation on-state voltage drop and a typical saturation on-state voltage drop, divided by the difference between a maximum saturation on-state voltage drop and a minimum saturation on-state voltage drop; the second drift rate is the difference between the on-state resistance and a typical on-state resistance, divided by the difference between a maximum on-state resistance and a minimum on-state resistance.
4. The screening method according to claim 1, wherein The screening method further comprises: Setting a screening grid coordinate, a drift amount as the X-axis of the screening grid coordinate, and a drift rate as the Y-axis of the screening grid coordinate; Marking the first drift rate and the first drift amount corresponding to each first switch tube into the screening grid coordinates; as well as The second drift rate and the second drift amount corresponding to each second switch tube are marked in the screening grid coordinates.
5. The screening method according to claim 4, characterized in that The screening grid coordinates include n first screening accuracies, m second screening accuracies, a first screening step and a second screening step, wherein the first screening step is composed of i first screening accuracies, and the second screening step is composed of j second screening accuracies, wherein i=1,2…n, j=1,2…m.
6. The screening method according to claim 5, characterized in that Wherein n is the minimum positive integer that satisfies a first inequality, wherein the first inequality is n×△x / 2≥max{|△Vth|}, wherein △x is the first screening accuracy, and max{|△Vth|} is a maximum value of the absolute values of all the first drift amounts and all the second drift amounts; m is the minimum positive integer that satisfies a second inequality, wherein the second inequality is m×△y / 2≥max{|△Vds|}, wherein △y is the second screening accuracy, and max{|△Vds|} is a maximum value of the absolute values of all the first drift rates and all the second drift rates.
7. The screening method according to claim 5, characterized in that The screening method further comprises: S21: In a grid corresponding to any one of the first screening accuracy in the first screening step and the second screening accuracy in the second screening step, confirm whether the grid contains at least one of the first switch tubes and at least one of the second switch tubes; S22: Continue screening the first switch tube and the second switch tube that are not in the grid in step S21, increase the number of the second screening precision of the grid to increase the second screening step of the grid, and confirm again whether the grid after increasing the second screening step has at least one of the first switch tube and at least one of the second switch tube; as well as S23: The number of the second screening precisions increases from 1 to m, wherein each time the number of the second screening precision increases by 1, step S22 is re-executed to obtain the grid having the largest second screening step length.
8. The screening method according to claim 7, characterized in that The screening method further comprises: S31: increasing the number of the first screening precision of the grid to increase the first screening step of the grid, and repeating step S21 and step S23; S32: the number of the first screening precision increases from 1 to n, and step S31 is executed once each time the number of the first screening precision increases by 1, so as to obtain the grid having the largest first screening step length; and S4: Determine the grid corresponding to any one of the second screening step lengths and / or any one of the first screening step lengths, and select the first switch tube and the second switch tube corresponding to the grid to be connected in parallel.
9. The screening method according to claim 6, characterized in that The first screening accuracy is a real number between 0 and the maximum value of the absolute values of all the first drift rates and all the second drift rates, and the second screening accuracy is a real number between 0 and 1; wherein the number of screening times of the screening method is inversely proportional to the first screening accuracy and the second screening accuracy.
10. The screening method according to claim 1, wherein The plurality of first switch tubes are insulated gate bipolar transistors, and the plurality of second switch tubes are wide bandgap semiconductor devices.