We will briefly introduce the concept of gate drivers and explain in detail the essential characteristics of isolated and non-isolated gate drivers. We will also learn about some key advantages of isolated gate drivers.
What is a gate driver?
In simple terms, a gate driver is a buffer circuit used to amplify low voltage or low current from a microcontroller or other source. In some cases, such as driving logic-level transistors used for digital signal transmission, using a microcontroller output does not compromise the application's efficiency, size, or thermal performance. However, in high-power applications, microcontroller outputs are generally unsuitable for driving high-power transistors.
But why use a microcontroller to drive power transistors? To better answer this question, let's consider a large-scale application. Switching power supplies are at the heart of almost every modern electrical system. Any device plugged into a wall outlet can utilize a switching power supply for power factor correction and generating DC current rails. Automotive systems use switching power supplies to power systems such as batteries, motors, and chargers. Grid infrastructure requires the efficient conversion of switching power supplied by DC solar panels to transfer electrical energy to DC storage systems and the AC grid.
Due to the increasing number and complexity of topologies in applications, modern switching power supplies typically use microcontrollers or other ASICs to coordinate the switching of high-power transistor arrays to meet precise switching timing requirements. This can be challenging because most microcontroller outputs are not optimized for driving power transistors.
High-power transistors (HPTs) exhibit characteristics almost entirely different from other transistors in analog signal chains or digital logic circuits. The breakdown voltage distribution of power transistors is extremely wide, ranging from approximately 40 volts to 1,200 volts or even higher. Due to the need for high-capacity drain circuitry and low on-state losses, drain-source resistances need to be as low as tens of milliohms or even less. The gate capacitance, inversely proportional to the drain-source resistance, typically exceeds 10,000 pF. Gate drive voltage and current requirements depend heavily on the transistor architecture and drain current ratings, with common values ranging from 8 to 30 volts and 1 to 5 amps. High-noise environments may even require bipolar output drives.
Compared to signal chains or digital transistors with frequencies of tens or hundreds of megahertz, traditional high-power transistors have a frequency ceiling of only a few hundred kilohertz, though this ceiling could potentially be pushed up by an order of magnitude with the advent of new technologies. This frequency limitation is due to increased gate capacitance and drive voltage requirements. The energy of a capacitor is equal to half the capacitance multiplied by the square of the voltage. The power consumption of charging and discharging the gate capacitor is equal to twice the capacitor's energy multiplied by the frequency—one charge, one discharge. A power transistor with a 15 nanofarad gate capacitance requires nearly half a watt of power consumption under 200 kHz, 12-volt square wave drive conditions. For converters capable of transmitting 3 to 5 kilowatts of power, the benefits of increasing the switching frequency, such as reducing the size and weight of the magnets, are sometimes more valuable than the cost of a few watts of drive losses.
Among the elements determining the driving requirements of a transistor, there is a more troublesome source of loss. During the charging and discharging of the gate capacitor, the switch experiences a transition period between the fully on and fully off states, during which a voltage appears on the switch and current flows through it. Due to the simultaneous presence of high voltage and high current, this type of switching loss results in considerable power consumption, sometimes reaching tens of watts, and further efficiency degradation. Therefore, it is beneficial to shorten the duration of the transition period by charging and discharging the gate capacitor more quickly.
If the output voltage is even high enough to turn on the transistor, then the low current signal provided by most microcontrollers will be ridiculously slow and extremely inefficient when driving high-power transistors.
Now let's answer the question of what a gate driver is. A gate driver is a circuit used to amplify control signals from a microcontroller or other source, thereby enabling the effective and efficient operation of a semiconductor switch.
Many gate drivers can operate under high bias voltages, such as those used in high-power converters.
Gate driver classification and advantages analysis
Broadly speaking, these gate drivers fall into two categories: non-isolated gate drivers and isolated gate drivers. Most non-isolated gate drivers used for operation at high voltages are half-bridge drivers.
Half-bridge drivers are designed to drive power transistors stacked together in a half-bridge configuration. They have two channels: a low-side and a high-side. The low-side is a fairly simple buffer, typically sharing the same ground point as the control input. The high-side, however, is carefully designed and referenced to the switching node of the half-bridge, allowing the use of two N-channel MOSFETs or two IGBTs. The switching node should transition quickly between the high-voltage bus and power ground, giving us the opportunity to cost-effectively power the high-side using the same power supply as when powering the low-side via a bootstrap circuit. To communicate whether the output should be high or low, a high-voltage level shifter must be included, which typically has low leakage current, only a few microamps or less.
This type of gate driver has several limitations. First, because it operates entirely on a single silicon die, it cannot exceed the limitations of silicon's process technology. Most non-isolated gate drivers operate at voltages not exceeding 700 volts. Second, level shifters must withstand the stress of high-voltage operation and must convey output states in high-noise environments. Therefore, level shifters typically incorporate some propagation delay to achieve adequate noise filtering. Then, the low-side driver must match the longer delay of the high-side driver. Third, non-isolated gate drivers for high-voltage operation are not flexible enough. Many complex topologies exist that require multiple outputs to be shifted above or below the control common level.
A feature increasingly common in modern gate drivers is the integration of an isolation layer between the input and output circuitry. These devices use one silicon die for control signals and another for output drive signals, physically isolating them by distance and insulating material. While control signals can pass through the isolation layer in various ways during transmission, unlike non-isolated gate drivers, the isolation layer prevents any significant leakage current from flowing from one side of the isolation layer to the other. Because one input die can be isolated from multiple output dies, and the output dies can be isolated from each other, the output common terminal can be freely offset upwards from the input common terminal or other output common terminals, up to the limits of the isolation technology.
Unlike non-isolated gate drivers, which have inflexible level shifters and predetermined output roles, isolated gate drivers can reference the output to any node in the circuit and can be configured as single-channel or dual-channel devices. The limits of isolation technology far exceed the silicon process limitations of non-isolated gate drivers, providing isolation layers with withstand voltages exceeding 5 kV. In addition to increased voltage limits and flexibility, isolated gate drivers can also enable faster, more robust operation. There are many reasons to use isolation. Many applications require isolated power supplies due to regulatory requirements, and isolated gate drivers can simplify system architecture. Sometimes, the strength of the isolation layer can also be used to enhance the system's resistance to surges, lightning strikes, and other abnormal events that could damage the system.
In other cases, the flexible use of isolation layers simplifies topology design, eliminating the need for signal converters or level shifters, such as inverting buck/boost converters. Even in traditional half-bridge applications where isolation is not critical, isolated gate drivers can outperform non-isolated gate drivers due to their superior propagation delay, higher drive force, and better tolerance to high-voltage transients.
Common topologies using isolated gate drivers include traction inverters, motor drivers, three-phase power factor correction circuits, and string photovoltaic inverters. These topologies all convert between AC and DC power supplies and connect directly to a high-voltage DC bus and a three-phase system such as a motor or power grid.
Disclaimer: This article is a reprint. If there are any copyright issues, please contact us promptly for deletion (QQ: 2737591964). We apologize for any inconvenience.
