Recent Advances in Short Circuit Protection Methods for SiC MOSFET

. Due to its high efficiency, low switching losses, and high temperature stability, Silicon Carbide (SiC) Metal-Oxide-Semiconductor Field-Effect Transistor (MOSFET) has been widely used in various fields. Therefore, the protection of SIC MOSFET devices is crucial. The significance of this research topic lies in comparing the existing short-circuit protection methods and technologies for SiC MOSFET, pointing out the advantages and shortcomings of each method, and providing suggestions. Firstly, the principles and impacts of HSF and FUL in short-circuit faults are introduced. Secondly, this article elaborates on short-circuit protection methods from two aspects: hardware protection and software protection, concentrating mostly on recently researched hardware protection methods. Finally, a comparison between the two protection methods is made, and potential improvement solutions are proposed. Based on the research conducted, the author believes that the current hardware protection method based on current detection is the optimal solution for SiC MOSFET short-circuit protection. Future research can be focused on integrating other solutions for further optimization on this basis.


Introduction
With the increasing demand for energy in modern society, energy conservation and emission reduction have become important issues in the world today.In this context, the demand for low-loss devices is becoming higher and higher, because low-loss devices can reduce energy waste and improve energy utilization efficiency.In the energy context, low-loss devices mainly refer to devices that can minimize energy loss during operation.Low-loss devices usually have characteristics such as low static power consumption, high efficiency, high speed, low switching loss, and high temperature stability.However, traditional Insulated Gate Bipolar Transistor (IGBT) devices gradually fail to meet some application requirements [1].
Silicon Carbide (SiC) Metal-Oxide-Semiconductor Field-Effect Transistor (MOSFET) is a silicon carbide-based metal-oxide semiconductor field-effect transistor.It has the advantages of high voltage, high frequency, high temperature, high power density, low on-resistance, and low switching loss [2].Due to these benefits, SiC MOSFET has a wide range of potential applications in the high-speed, high-power, and high-temperature areas.For example, SiC MOSFET can replace traditional IGBT devices in high-speed railways [3], improve railway efficiency and operating speed under high temperature and high voltage environments; it can be applied in new energy vehicles [4], effectively reduce the weight and volume of electric vehicle batteries, improve battery energy density and extend battery life; it can also be used for power control and drive of industrial automation equipment [5], improving equipment efficiency and stability.The high-frequency switching capability and low switching resistance of SiC MOSFET also make it an ideal choice for solar inverters and high-frequency power supplies [6].
However, SiC MOSFET has several drawbacks, the most significant of which is its weak capacity to withstand short-circuit current, making it susceptible to short-circuit errors.The short-circuit current is the highest current that, in the event of a short circuit, can harm a SiC MOSFET.Therefore, it is crucial to research SiC MOSFET's short-circuit protection mechanism.
Although SiC MOSFETs possess better thermal conductivity than Si devices, achieving effective shortcircuit protection is more challenging with SiC MOSFETs compared to Si devices [7].First, SiC MOSFETs have a higher current density and a smaller chip area compared to other MOSFETs with the same rated current capacity.However, this characteristic also reduces their capability to withstand short-circuits.Second, In case the interface quality of SiC MOSFET is not up to the mark, it may result in problems with the dependability of the gate oxide layer when short-circuit situations occur.SiC MOSFETs demand a larger forward gate bias, hence an increase in the gate electric field will make the problem of the gate oxide layer degrading during short-circuits even worse.Because SiC MOSFETs have insufficient short-circuit resistance, the protection circuit needs to respond faster to ensure safe and reliable operation within the allowable operating range.However, compared with Si devices, SiC MOSFET has smaller junction capacitance and faster switching speed.In addition, the unique positive temperature coefficient transconductance of SiC MOSFET leads to the increase of dI/dt and dV/dt when it is turned on with the increase of junction temperature.When subjected to higher dI/dt and dV/dt conditions, The circuit that provides protection against short-circuit events for SiC MOSFETs must possess both fast response and noise resistance, which can be challenging to maintain in equilibrium.
Currently, how the SiC MOSFET is safeguarded against short-circuiting primarily comprises of two approaches: hardware protection and software protection.Software protection refers to protecting SiC MOSFET by controlling its switching time and frequency, etc.For example, adding short-circuit detection circuits and short-circuit protection algorithms to the control circuit to achieve software protection.
This article summarizes the research progress on the protection mechanism for preventing short circuits in SiC MOSFET in recent years, mainly including software protection and hardware protection, and provides an outlook on the development of future Safeguards against short-circuit events.

Short-circuit failure of SiC MOSFETs
Based on the short-circuit loop inductance value and location, Short-circuits in SiC MOSFETs can be divided into two categories, namely Class 1 and Class 2, as shown in Table 1 [8].The circuit inductance is in the μm range

Class 1 short-circuit
Class 1 short-circuit pertains to a situation where the circuit has low inductance (below 10nH) and is located near the SiC MOSFET, resulting in a rapid surge in current.SiC MOSFET class 1 short circuits can be categorized into two types based on the timing of their occurrence: hard-switching faults (HSF) and load faults (FUL).The short circuit test circuit is shown in Figure 1(a).
Hard-switching fault refers to a fault phenomenon that occurs in SiC MOSFET during the switching process.That is, during switching, the conduction and cutoff states of the MOSFET switch suddenly, causing high peak voltage and current in the circuit.This creates significant instantaneous pressure and damage to the MOSFET and other circuit components, and can even cause failure or damage.When HSF occurs, the drain current ID rapidly rises to its maximum value and then falls back to a stable short-circuit current value.Due to the small loop inductance, the drain-source voltage VDS drops slightly and then stabilizes at the bus voltage [8].The typical waveform of this fault is shown in Figure 1(b).
Load fault refers to a type of fault in which the entire circuit system cannot operate normally due to problems with the load when a load is connected to the output port of SiC MOSFET.Common load faults include short circuits, open circuits, the effects of inductance and capacitance, overload, etc. Load faults may cause errors in current, voltage, power, or distorted waveforms, and also increase the working temperature of SiC MOSFET, thereby affecting its reliability and lifespan.In the event of a failure under load (FUL), the voltage on both ends of the SiC MOSFET rises to the bus voltage, leading to a sudden surge in short-circuit current from the load current [8].The typical waveform when the fault occurs is shown in Figure 1(c).

Class 2 short-circuit
Class 2 short circuit refers to a momentary short circuit fault in the circuit, usually referring to a short circuit phenomenon with a large inductance (above a certain level) and a short circuit location far from the SiC MOSFET.In contrast to Class 1 short circuit, Class 2 short circuit has a slower increase rate of short circuit current, lower instantaneous power, and gradual increase of short circuit current within a specific time frame, without a sudden impact on the SiC MOSFET.As a result, the channel resistance of the SiC MOSFET does not experience a sudden increase, which prevents immediate partial voltage loss and has minimal impact on the overall reliability of the SiC MOSFET.3 Short-circuit protection for SiC MOSFET

Short-circuit protection method based on current detection
A typical form of short circuit prevention is electric current detection, which monitors changes in current to protect SiC MOSFETs from short circuits.Typically, a current detection resistor is inserted in the circuit in series to monitor the circuit current.Once the current surpasses the predetermined threshold, the circuit protection measures are activated; a high-precision current detection circuit can be used to detect zero drift of the current, which can help determine if there is a short circuit.Existing research has proposed a method of protecting SiC MOSFET by monitoring the current through a current sensor.When the current exceeds a certain threshold, the circuit will turn off the SiC MOSFET through a switch controller.This can prevent SiC MOSFET from overcurrent and damage.
Some researchers proposed a protection method based on tunnel magnetoresistance (TMR) for SiC MOSFET short circuit protection in HybridPACK™ driver packages [9], as shown in the schematic diagram in Figure 3.One approach is to install TMR sensors at both ends of the MOSFET for real-time detection of the magnetic field and current.This enables quick identification of any short-circuit faults and allows for immediate disconnection of the MOSFET's power supply.By introducing sensors for real-time monitoring and protection, SiC MOSFET short-circuit faults can be quickly and accurately detected, avoiding damage caused by short-circuits, improving system reliability and stability, and with less susceptibility to circuit board noise and interference, resulting in higher stability.The specific implementation circuit is depicted in the figure 4.However, this short-circuit protection method also has some drawbacks, such as the power consumption of the current sensor, which can have a certain impact on the energy efficiency of the system.In addition, the accuracy of detecting and protecting SiC MOSFET short-circuit faults is limited by the accuracy and resolution of the current sensor.
Some researchers have proposed to protect circuits by detecting high currents and implementing mixed shutdown [10], as shown in Figure 5.This method can reduce peak currents and voltages while protecting the circuit, but requires reasonable design and control to ensure its reliability and stability.

Short-circuit protection method based on voltage detection
The voltage detection method is another commonly used short circuit protection method, which limits the short circuit current by detecting the drain-source voltage variation of SiC MOSFET.Compared with the current detection method, the voltage detection method has lower implementation difficulty, but its detection error is larger, and it needs to be reduced through circuit calibration and other methods.
Some researchers proposed a SiC MOSFET shortcircuit protection circuit based on leakage source voltage integration [11], as shown in Figure 6.This circuit integrates the leakage source voltage to obtain the time integral value of the leakage source voltage.When the integral value exceeds a certain threshold, the short-circuit protection circuit is triggered to turn off the SiC MOSFET, preventing it from being damaged.This method can monitor the leakage-source voltage in real time, but due to the need to integrate the leakage-source voltage signal at multiple moments to determine whether a short circuit occurs, the time is longer and cannot effectively prevent damage to the SiC MOSFET.Therefore, further research is needed in this area.
Researchers have also discussed the current sharing and overvoltage issues in parallel silicon carbide MOSFET modules, and proposed a voltage-controlled short-circuit protection method.
In the parallel design of silicon carbide MOSFET modules, current sharing is a very important issue.If one MOSFET in the module fails or is damaged, it may cause other MOSFETs to bear too much load current, which may lead to the failure of the entire module.Similarly, overvoltage issues can also cause module failure, as SiC MOSFETs are more sensitive to high voltages.
To address these issues, researchers have proposed a voltage-controlled short-circuit protection method [12].This method utilizes the voltage difference between each MOSFET in the parallel module to achieve current allocation and short-circuit protection by controlling the voltage of the module.Specifically, the output end of the parallel module is connected to a voltage sensor to determine the current load of each MOSFET by sensing the voltage difference between them.When the current of a certain MOSFET is too high, its voltage will also increase accordingly.At this point, the control circuit will adjust and lower the voltage of the MOSFET through PWM technology to achieve balanced current allocation.In case of a short circuit, the control circuit will reduce the voltage of all MOSFETs in unison to achieve short-circuit protection.The gate drive circuit topology is shown in Figure 7.This method has the advantages of simple implementation, low cost, and good performance.However, it should be noted that strict circuit design, parameter calculation, and experimental verification are required in practical applications to ensure its reliability and stability.

Other detection-based short-circuit protection methods
In addition to current and voltage detection, there are also other methods for short-circuit protection.For example, some researchers have proposed using power indirect dissipation level as a reference value for shortcircuit protection [13].By comparing the power indirect dissipation level during a short circuit with a preset threshold, short-circuit protection can be achieved.The circuit for the proposed IPDL-SCP method is shown in Figure 8.This method has advantages such as being fast, reliable, and highly adaptable.Compared to traditional current suppression-based short-circuit protection methods, this method can quickly and accurately detect short-circuit situations and take immediate action, avoiding the problem of delayed response.However, this method also has certain limitations.The detection and judgment of short-circuit faults are based on indirect power dissipation levels, which are affected by many factors such as load changes and temperature fluctuations, which may result in misjudgment.This method may have a negative impact on the performance and stability of the equipment because the cutoff operation triggered by short-circuit protection may affect the normal operation of the equipment, especially in high-power output applications.

Software Protection
Software protection is achieved by adjusting the parameters or algorithms of the control system to protect the SiC MOSFET device.This method has advantages such as low implementation cost and easy debugging compared to hardware methods, but it also has some disadvantages, such as slow response due to problems with parameter settings.

Model-predictive-based short-circuit protection method
This approach utilizes the SiC MOSFET model for forecasting and accomplishes the identification and safeguarding of short-circuit faults by evaluating and interpreting model parameters.The advantages of this method are that it does not require the addition of sensors or other hardware, reduces hardware costs and system complexity, and can predict circuit faults before they occur, avoiding circuit damage.However, the performance of the model is limited by the model parameters.If the parameters are too low, the predictive ability is poor and cannot effectively protect the circuit; if the parameters are too high, it will increase the system calculation and operation time, resulting in a slower response and affecting the circuit protection.This method needs to consider factors such as model establishment and optimization, calculation and operation time, and needs to be selected and optimized based on specific application scenarios.

Feedback-based short-circuit protection method
By implanting a feedback control algorithm program into the controller, the current and voltage signals of SiC MOSFET in the circuit are constantly monitored and analyzed.Upon the occurrence of a short-circuit fault, the controller instantly adapts the current and voltage parameters of SiC MOSFET to attain the optimal protection state, thereby preventing the SiC MOSFET from being damaged.

Comparison between Hardware Protection and Software Protection
Table 2 summarized the comparisons between different protection strategies.The advantage of hardware protection lies in its ability to quickly detect excessive current or short circuits, and immediately stop the power supply through the protection circuit to protect SiC MOSFET from damage.Over-current protection can usually be customized to determine the protection threshold based on specific application requirements, reducing the risk of false protection.Compared to software protection, hardware protection has a faster time response.However, hardware protection usually increases circuit complexity, occupies additional space and cost, and usually requires additional circuit energy consumption, which may affect circuit efficiency and reduce system reliability.For these disadvantages, some optimization solutions can be adopted, such as using simpler protection circuits, reducing the protected area, using more efficient protection components, etc., to reduce the complexity of the protection circuit.Nowadays, there are many specialized protection chips on the market that can help designers achieve shortcircuit protection for SiC MOSFET.These chips usually integrate multiple protection functions, and they can reduce circuit complexity and improve protection effectiveness.
Software protection can provide more flexible protection mechanisms while protecting SiC MOSFET.For example, the threshold can be configured and environment awareness can be provided.And usually, it can more finely monitor changes in parameters such as current and voltage, and provide timely feedback and adjustment.However, in some cases, the response time of software protection may be longer than that of hardware protection.This may lead to adverse consequences in the case of transient current surges into SiC MOSFET.In addition, software protection is more dependent on the normal operation of the external environment, such as the normal operation of the CPU and the execution of software logic, which are necessary prerequisites for the normal operation of software protection.As software protection is usually implemented on the processor, the complexity of protective hardware may be different, and the protective performance implemented on hardware may be better than that of software.To overcome these shortcomings, the design of the controller can be optimized, faster controller chips and more efficient algorithms can be used to improve the response speed of software protection, or predictive control algorithms can be used to predict the occurrence of short-circuit events in advance based on the working status of the circuit, load changes, and other information, so that measures can be taken to protect the circuit before a short-circuit event occurs, thereby shortening the response time.

Prospects
The author has the following outlook on future shortcircuit protection methods for SiC MOSFET: 1) Improve protection efficiency: Although the currently used short-circuit protection methods can protect SiC MOSFET from short-circuit damage, they have the problems of large protection delay and insufficient protection capability.Future research should focus on improving protection efficiency and reducing protection delay by optimizing the structure and algorithm of the protection circuit to improve protection speed and accuracy.2) Improve reliability: As a new type of semiconductor device, SiC MOSFET still have some issues in terms of lifespan and reliability.The focus should be on how to improve the reliability of SiC MOSFET, including reducing device damage risks, improving manufacturing processes, and optimizing packaging structures.3) Research on new protection methods: In addition to conventional short-circuit protection methods, new protection methods should also be explored, such as protection algorithms based on machine learning and adaptive protection algorithms, to improve protection efficiency and accuracy.
Research on integrated solutions for short-circuit protection and power electronic system control: Explore integrated solutions for short-circuit protection and power electronic system control to improve the overall performance and reliability of the system.

Conclusion
Short circuit safeguarding for SiC MOSFET has become an important research field.Hardware-based short-circuit protection techniques offer benefits such as swift response time and enhanced dependability, but they also pose some drawbacks, including intricate design and elevated expenses.Software short-circuit protection methods have advantages such as low implementation cost and convenient debugging compared to hardware methods, but also have some disadvantages, such as slow response speed.Future research should be conducted indepth from both hardware and software aspects to find the optimal short-circuit protection method for SiC MOSFETs.

Table 1 .
Characteristics of short-circuit types.

Table . 2
Comparison between Hardware Protection and Software Protection