A half-bridge adaptive dead-time control circuit based on switching node voltage detection
By using a half-bridge adaptive dead-time control circuit based on switching node voltage detection, the dead time is dynamically adjusted, solving the problem that the dead time in the half-bridge structure cannot be adaptive, and realizing safe operation and efficiency improvement of power devices.
Patent Information
- Application Number
- CN202611536845.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-30
- Publication Date
- 2026-11-17
AI Technical Summary
In existing half-bridge structures, the dead time cannot be adaptively adjusted according to the type of power device, switching speed, load size, and operating conditions, resulting in problems such as increased freewheeling loss, device heating, reduced efficiency, and electromagnetic interference. Existing solutions have limitations and are costly.
A half-bridge adaptive dead-time control circuit based on switching node voltage detection is adopted. By connecting the lower bridge drive circuit, SW valley detection circuit, pre-dead-time adaptive circuit and post-dead-time reset circuit in series, the actual turn-off time of the upper bridge power device is determined by directly detecting the switching node voltage and dynamically adjusting the dead time.
It effectively eliminates the freewheeling loss, heat generation, and electromagnetic interference caused by excessive dead time in the body diode, improves conversion efficiency, avoids shoot-through between the upper and lower bridges, and is suitable for synchronous rectification, synchronous freewheeling, and motor drive applications.
Smart Images

Figure CN123122224A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of semiconductor technology, and more specifically, to a half-bridge adaptive dead-time control circuit based on switching node voltage detection. Background Technology
[0002] With the development of the power semiconductor industry and the widespread application of various power semiconductors in power supplies and electric drives, half-bridge topologies have been widely used in switching power supplies, synchronous rectifier converters, synchronous freewheeling converters, and motor drive inverters. In a half-bridge structure, a dead time must be set between the turn-on and turn-off of the upper and lower bridge power devices to prevent shoot-through and damage. However, the optimal dead time for a power half-bridge varies depending on the type of power device, switching speed, load size, load change rate, and operating conditions: an excessively long dead time will cause the body diode of the lower bridge power device to freewheel for an extended period, leading to increased freewheeling losses, device heating, decreased conversion efficiency, and causing waveform oscillations and electromagnetic interference at the switching node; an excessively short dead time poses a risk of shoot-through between the upper and lower bridges. Therefore, enabling the dead time to adaptively adjust according to the characteristics of the power devices, load conditions, and operating conditions is an essential requirement that the half-bridge drive technology field urgently needs to meet.
[0003] Currently, to address the issue of dead time in traditional half-bridge structures not being perfectly adapted due to variations in load size, power device switching speed, and load change rate, two main mature solutions exist in the market. The first is the pre-calibration and correction scheme: This involves proactively testing the dead time under different power devices, switching speeds, loads, temperatures, and operating conditions, and pre-writing the test results into a program for dead time correction during practical applications. While this is the mainstream approach in the current market, it is significantly limited by the finite number of pre-test samples, making it difficult to cover the complex and varied operating conditions in real-world applications. The correction error is relatively large, and the pre-testing and parameter calibration work is cumbersome and labor-intensive. The second solution is the microprocessor-based intelligent acquisition scheme: This uses a microprocessor to intelligently acquire information such as the dead time and current troughs of each half-bridge, dynamically adjusting the dead time accordingly. This scheme requires high-speed analog-to-digital conversion voltage sampling, zero-crossing detection, and current detection, resulting in large size, high cost, and high technical difficulty, making it unsuitable for a wide range of applications.
[0004] Therefore, there is an urgent need for an adaptive dead-time control circuit, device, and chip suitable for half-bridge structures composed of power devices such as MOSFETs, IGBTs, and silicon carbide. This circuit should be able to determine the dead time in real time based on the actual voltage state of the half-bridge switching nodes without relying on prior calibration data and high-speed sampling by the microprocessor. This would effectively solve problems such as power device damage, overheating, efficiency reduction, and electromagnetic interference caused by improper dead time, and could be widely applied in power supply and motor drive fields containing half-bridge topology components, such as half-bridge and full-bridge circuits. Summary of the Invention
[0005] The purpose of this application is to overcome the shortcomings of existing technologies and provide a half-bridge adaptive dead-time control circuit based on switching node voltage detection. This circuit can be widely used in power supplies, motor drives and other fields with various half-bridge or similar topologies, effectively solving various application problems caused by improper dead time under various operating conditions.
[0006] The objective of this application is achieved through the following technical solution: In the first aspect, this application proposes a half-bridge adaptive dead-time control circuit based on switching node voltage detection, which is connected in series between the lower bridge drive output terminal of the external half-bridge gate drive circuit and the gate of the lower bridge power device of the power half-bridge. It includes an input port Lin, an output port Lout, a sampling port SW, a power supply terminal VCC and a ground terminal GND, as well as a lower bridge drive circuit 11, an SW valley detection circuit 14, a pre-dead-time adaptive circuit 12 and a post-dead-time reset circuit 13. The input port Lin is used to connect to the lower bridge drive output of the external half-bridge gate drive circuit and receive the lower bridge gate control signal. The lower bridge gate control signal has an on command level and an off command level. The lower bridge drive circuit 11 includes a pull-up switch, a pull-down switch, and a push-pull control node. The pull-up switch is connected between the power supply terminal VCC and the output port Lout, and the pull-down switch is connected between the output port Lout and the ground terminal GND. The control terminals of the pull-up switch and the pull-down switch are connected to form a push-pull control node. The output port Lout is used to connect to the gate of the lower bridge power device. When the push-pull control node is at the first control level, the pull-up switch is turned on and the pull-down switch is turned off, and the output port Lout outputs an on level to turn on the lower bridge power device. When the push-pull control node is at the second control level, which is the opposite of the first control level, the pull-up switch is turned off and the pull-down switch is turned on, and the output port Lout outputs an off level to turn off the lower bridge power device and provides a charge discharge circuit for the gate of the lower bridge power device. The SW valley detection circuit 14) includes a comparator. The first input terminal of the comparator is connected to the sampling port SW via the input branch. The sampling port SW is used to connect to the switching node of the power half-bridge. The second input terminal of the comparator is connected to the reference voltage. When the voltage of the switching node is lower than the reference voltage, the output terminal of the comparator flips from an invalid level to an effective level and outputs a valley signal. The input terminal of the post-dead-time reset circuit 13) is connected to the input port Lin, which is used to generate a valid reset signal when the lower bridge gate control signal changes from the turn-on command level to the turn-off command level, and the reset signal becomes invalid after the turn-off command level appears and continues to be held for a preset reset duration. The pre-dead-zone adaptive circuit 12 includes a trigger unit and a confirmation unit; the set input terminal of the trigger unit is connected to a fixed valid level, the asynchronous reset terminal of the trigger unit is connected to the output terminal of the post-dead-zone reset circuit 13 to receive a reset signal, and the output terminal of the trigger unit is connected to a push-pull control node; the confirmation unit is connected between the output terminal of the SW valley detection circuit 14 and the clock terminal of the trigger unit, and is used to output a set trigger edge to the clock terminal of the trigger unit after the valley signal has been valid for a preset confirmation time. When the reset signal is invalid and a set trigger edge appears at the clock terminal of the trigger unit, the output terminal of the trigger unit sets the push-pull control node to the first control level to turn on the lower bridge power device; when the reset signal is valid, the trigger unit is asynchronously reset, and the output terminal of the trigger unit sets the push-pull control node to the second control level to turn off the lower bridge power device.
[0007] The main solution and its various further alternatives described above can be freely combined to form multiple solutions, all of which are solutions that can be adopted and are claimed in this application; the alternatives can also be freely combined with each other and with other alternatives. Those skilled in the art, after understanding the solution of this application, will realize from the prior art and common general knowledge that there are many combinations, all of which are technical solutions to be protected in this application, and will not be exhaustively listed here.
[0008] This invention discloses a half-bridge adaptive dead-time control circuit based on switching node voltage detection, which is connected in series between the lower bridge drive output terminal of the external half-bridge gate drive circuit and the gate of the lower bridge power device of the power half-bridge. It includes a lower bridge drive circuit, an SW valley detection circuit, a pre-dead-time adaptive circuit and a post-dead-time reset circuit, as well as an input port Lin, an output port Lout, a sampling port SW, a power supply terminal VCC and a ground terminal GND. The lower bridge drive circuit consists of a push-pull output stage composed of pull-up and pull-down switches. When the lower bridge is turned off, the pull-down switch provides a charge discharge circuit for the gate of the lower bridge power device. The SW valley detection circuit compares the voltage of the power half-bridge switching node with the reference voltage and outputs a valley signal when the switching node voltage is lower than the reference voltage. The pre-dead-time adaptive circuit generates a set trigger edge after the valley signal has been valid for a preset confirmation time, driving the push-pull control node to the on level and turning on the lower bridge power device in advance. The post-dead-time reset circuit generates a reset signal when the lower bridge gate control signal changes from the on command level to the off command level, causing the trigger unit to be reset asynchronously and the lower bridge power device to be turned off synchronously. The reset signal becomes invalid after a preset reset holding time.
[0009] This invention determines the actual turn-off time of the upper bridge power device by directly detecting the voltage of the switching node, thus adaptively shortening the dead time according to the load, device switching speed, and operating conditions. It eliminates the need for high-speed sampling and pre-stored correction parameters by a microprocessor, effectively eliminating the risks of body diode freewheeling loss, heat generation, waveform oscillation, and false conduction caused by excessive dead time. At the same time, it avoids shoot-through between the upper and lower bridges and is widely applicable to power supply and drive fields containing half-bridge topologies, such as synchronous rectification, synchronous freewheeling, and motor drives. Attached Figure Description
[0010] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0011] Figure 1 A schematic diagram of a half-bridge adaptive dead-time control circuit based on switching node voltage detection proposed in an embodiment of this application is shown.
[0012] Figure 2 The logic timing diagram of the circuit is shown.
[0013] Figure 3 A schematic diagram of an adaptive dead-time control circuit for a half-bridge structure of a power device is shown.
[0014] Figure labeling: 11-Lower bridge drive circuit; 12-Pre-dead-time adaptive circuit; 13-Post-dead-time reset circuit; 14-SW valley detection circuit; Lin-Input port; Lout-Output port; SW-Sampling port; VCC-Power supply terminal; GND-Ground terminal. Detailed Implementation
[0015] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.
[0016] Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0017] Traditional half-bridge structures cannot perfectly adapt their dead time due to characteristics such as load size, power device switching speed, and load change rate. Current mature solutions rely on microprocessors to intelligently collect the dead time and current troughs of each half-bridge, which is characterized by large size, high cost, and high technical difficulty. This application proposes a half-bridge adaptive dead time control circuit based on switching node voltage detection. It can intelligently detect the half-bridge phase line voltage and, through built-in triggers, trough samplers, and logic circuits, adaptively adjust the dead time of traditional PWM control or the built-in dead time of the half-bridge drive. When the half-bridge phase line voltage falls below a certain amplitude, after a minimum necessary dead time, the internal logic circuit can control the lower bridge to turn on directly. This significantly improves the efficiency of synchronous rectification, synchronous freewheeling, and other applications, and alleviates oscillations, false triggering, and overheating caused by excessive dead time. Furthermore, it can be widely applied to various half-bridge or similar topologies in power supplies and motor drives, effectively solving problems such as low efficiency, overheating, high failure rate, power device waveform oscillations, and EMI caused by unsuitable dead time.
[0018] Figure 1This illustration shows a schematic diagram of a half-bridge adaptive dead-time control circuit based on switch node voltage detection according to an embodiment of this application. This embodiment provides a half-bridge adaptive dead-time control circuit based on switch node voltage detection, which is connected in series between the lower bridge drive output terminal of the external half-bridge gate drive circuit and the gate of the lower bridge power device of the power half-bridge. That is, this circuit is inserted into the existing lower bridge drive path, and after adaptive dead-time processing of the lower bridge gate control signal output by the external half-bridge gate drive circuit, it drives the lower bridge power device of the power half-bridge. The circuit includes five external ports: input port Lin, output port Lout, sampling port SW, power supply terminal VCC, and ground terminal GND, as well as four functional modules: lower bridge drive circuit 11, SW valley detection circuit 14, pre-dead-time adaptive circuit 12, and post-dead-time reset circuit 13. The input port Lin is used to connect to the lower bridge drive output of the external half-bridge gate drive circuit and receive the lower bridge gate control signal. The output port Lout is used to connect to the gate of the lower bridge power device. The sampling port SW is used to connect to the switching node of the power half-bridge. The power supply terminal VCC and the ground terminal GND are used to supply power to the circuit. The lower bridge gate control signal has two states: turn-on command level and turn-off command level, which are used to represent the turn-on command and turn-off command issued by the external half-bridge gate drive circuit, respectively.
[0019] Figure 3A schematic diagram of an adaptive dead-time control circuit for a power device half-bridge structure is shown. During one switching cycle of the power half-bridge, when the external half-bridge gate drive circuit outputs a turn-off command and the lower bridge gate control signal changes from an on command level to a turn-off command level, the post-dead-time reset circuit 13 generates a valid reset signal. The trigger unit is asynchronously reset, and its output makes the push-pull control node a second control level. The pull-up switch is turned off, the pull-down switch is turned on, and the output port Lout outputs a turn-off level to turn off the lower bridge power device and provide a charge discharge circuit for the gate of the lower bridge power device. After that, the reset signal becomes invalid after a preset reset holding time, releasing the asynchronous reset of the trigger unit. When the external half-bridge gate drive circuit subsequently turns off the upper-bridge power device and the lower-bridge power device's body diode conducts freewheeling, the voltage of the switching node drops below the reference voltage Vref. The comparator CMP of the SW valley detection circuit 14 outputs a valley signal. After the valley signal remains valid for a preset confirmation time, the confirmation unit outputs a set trigger edge to the clock terminal of the trigger unit. At this time, since the reset signal has become invalid, the trigger unit is set, and its output terminal sets the push-pull control node to the first control level, turning on the pull-up switch and turning off the pull-down switch. The output port Lout outputs an on-level, thus turning on the lower-bridge power device before the fixed dead time set by the external half-bridge gate drive circuit ends. Conversely, if a set trigger edge appears at the clock terminal of the trigger unit while the reset signal is still valid, the trigger unit is asynchronously reset when the reset signal is valid, and its output terminal sets the push-pull control node to the second control level to turn off the lower-bridge power device. Therefore, the lower-bridge power device will not be turned on. Therefore, this embodiment determines the actual turn-off time of the upper bridge power device by directly detecting the actual voltage state of the power half-bridge switching node, so that the dead time is adaptively shortened according to the power device type, switching speed, load size and operating conditions. There is no need to pre-test and write correction parameters, nor is there a need for the microprocessor to perform high-speed analog-to-digital sampling and current detection, thereby effectively eliminating problems such as power device damage, heat generation, efficiency reduction, waveform oscillation and electromagnetic interference caused by improper dead time.
[0020] Figure 2The logic timing diagram of the circuit is shown. The SW trough detection circuit 14 includes a comparator CMP, a current-limiting resistor Rsw, a clamping voltage regulator branch, and a filter capacitor C2. The current-limiting resistor Rsw and the clamping voltage regulator branch together form the input branch connecting the sampling port SW and the first input terminal of the comparator CMP. Specifically, the phase line of the external power half-bridge is introduced through the input port SW. The sampling port SW is connected to the first terminal of the clamping voltage regulator branch via the current-limiting resistor Rsw. The second terminal of the clamping voltage regulator branch is connected to ground GND. The first terminal of the clamping voltage regulator branch is connected to the first input terminal of the comparator CMP. The second input terminal of the comparator CMP is connected to the reference voltage Vref. The filter capacitor C2 is connected between the first input terminal of the comparator CMP and ground GND. Thus, the switching node voltage sampled by the sampling port SW is conditioned through current limiting, clamping, and filtering before being sent to the comparator CMP for comparison with the reference voltage Vref.
[0021] A current-limiting resistor Rsw is connected in series between the sampling port SW and the first terminal of the clamping voltage regulation branch to limit the current flowing into the internal circuit from the sampling port SW. Since the voltage amplitude of the switching node of the power half-bridge depends on its bus voltage level, the voltage amplitude experienced by the sampling port SW varies greatly in applications with different bus voltage levels. Therefore, the resistance value of the current-limiting resistor Rsw is selected according to the bus voltage level of the power half-bridge: the higher the bus voltage level, the larger the resistance value of the selected current-limiting resistor Rsw, thereby confining the current flowing into the internal circuit within a safe range; when the bus voltage level is low, the resistance value of the current-limiting resistor Rsw is correspondingly reduced to ensure detection sensitivity. In this embodiment, the current-limiting resistor Rsw is located outside the chip and configured by the application end according to the bus voltage level, thus enabling the same circuit to be used in both low-voltage and high-voltage applications.
[0022] The clamping voltage regulation branch consists of the fourth resistor R4 and the first Zener diode D1. The first end of the fourth resistor R4 is connected to the end of the current-limiting resistor Rsw furthest from the sampling port SW, serving as the first terminal of the clamping voltage regulation branch; the second end of the fourth resistor R4 serves as the second terminal of the clamping voltage regulation branch and is connected to ground GND. The cathode of the first Zener diode D1 is connected to the first terminal of the clamping voltage regulation branch, and the anode of the first Zener diode D1 is connected to ground GND. The first Zener diode D1 is a high-power Zener diode, which, together with the fourth resistor R4, forms an internal voltage regulation structure. When the switching node voltage is the bus voltage, the injected current, after being limited by the current-limiting resistor Rsw, is mainly discharged through the reverse breakdown path of the first Zener diode D1. This clamps the potential of the first terminal of the clamping voltage regulation branch to be near the Zener value of the first Zener diode D1, thereby converting the high voltage swing of the switching node into an acceptable level in the internal low voltage domain, preventing the high voltage from being directly applied to the first input terminal of the comparator CMP and damaging the internal components. The fourth resistor R4 is connected between the first terminal of the clamping voltage regulation branch and the ground terminal GND, providing a static working path for the first Zener diode D1 and forming a discharge circuit when the switching node voltage is low. This allows the potential of the first terminal of the clamping voltage regulation branch to decrease as the switching node voltage actually decreases, thereby ensuring that the comparator CMP can accurately reflect the changes in the switching node voltage.
[0023] In this embodiment, the value of the current-limiting resistor Rsw is selected based on the bus voltage level of the power half-bridge, ensuring that the operating current of the first Zener diode D1 when the voltage at the switching node is equal to the bus voltage is between the minimum stable breakdown current and the maximum permissible breakdown current of the first Zener diode D1. The minimum stable breakdown current is the minimum current required to ensure that the first Zener diode D1 reliably enters the voltage regulation region and stabilizes the clamping potential. If the operating current is lower than this value, the first Zener diode D1 will exit the voltage regulation region, and the clamping potential will drift with the bus voltage, leading to distorted detection results. The maximum permissible breakdown current is the maximum current that the first Zener diode D1 can withstand for an extended period without overheating. If the operating current is higher than this value, the first Zener diode D1 will be damaged due to excessive power consumption. Therefore, in low-voltage applications, a smaller value of the current-limiting resistor Rsw can ensure that the operating current falls within the aforementioned range. In high-voltage applications, the value of the current-limiting resistor Rsw is increased accordingly, thus allowing the circuit to adapt to different bus voltage levels while maintaining the same clamping voltage regulation branch structure.
[0024] In this embodiment, the reference voltage Vref is set to 2.5V. For unconventional conditions such as oscillations and noise present at typical switching nodes, the ringing trough is much higher than 2.5V and will not fall below 2.5V. Only when the upper-bridge power device is completely turned off and the lower-bridge power device freewheels will the voltage of the switching node drop below 2.5V, thereby triggering the comparator CMP. Therefore, using 2.5V as the comparison threshold can distinguish the actual freewheeling trough from the ringing trough in terms of amplitude, avoiding misjudging ringing as upper-bridge shutdown.
[0025] The filter capacitor C2 is connected between the first input terminal of the comparator CMP and the ground terminal GND. It uses the high-frequency characteristics of ringing and noise to filter them, isolating ringing, oscillation and noise under abnormal conditions, forming a second discrimination barrier in the time / frequency dimension. The amplitude discrimination and filtering discrimination work together to ensure that the comparator CMP outputs a valid trough signal only when the switching node actually enters the freewheeling state, fundamentally avoiding the risk of shoot-through between the upper and lower bridges caused by false triggering.
[0026] In this embodiment, the lower bridge gate control signal is provided by the lower bridge drive output terminal of the external half-bridge gate drive circuit and input through the input port Lin. Its turn-on command level is high and its turn-off command level is low. That is, when the input port Lin is high, it indicates that the external half-bridge gate drive circuit has issued an turn-on command, and when the input port Lin changes from high to low, it indicates that the external half-bridge gate drive circuit has issued a turn-off command. Correspondingly, the output level of the output port Lout is also defined in a high-level active manner. Its turn-on level is high and its turn-off level is low. That is, when the output port Lout outputs a high level, the lower bridge power device is turned on, and when the output port Lout outputs a low level, the lower bridge power device is turned off, so that the output polarity of the output port Lout is consistent with the drive polarity required by the gate of the lower bridge power device. The lower bridge drive circuit 11 adopts a push-pull output stage structure, with a PMOS pull-up switch Q2 and an NMOS pull-down switch Q2. The source of the PMOS pull-up switch Q2 is connected to the power supply terminal VCC and the drain is connected to the output port Lout. The source of the NMOS pull-down switch Q2 is connected to the ground terminal GND and the drain is connected to the output port Lout. The gate of the PMOS pull-up switch Q2 and the gate of the NMOS pull-down switch are connected to form a push-pull control node. The output port Lout is connected to the gate of the lower bridge power device through a gate drive resistor. Since the gates of the PMOS pull-up transistor Q2 and the NMOS pull-down transistor are connected to the same node and the gate control levels required for their conduction are opposite, when the push-pull control node is at a low level (first control level), the PMOS pull-up transistor Q2 is turned on and the NMOS pull-down transistor is turned off. The power supply terminal VCC supplies power to the output port Lout through the PMOS pull-up transistor Q2, and the output port Lout outputs a high level. This high level then charges the gate of the lower bridge power device through the gate drive resistor, turning on the lower bridge power device. When the push-pull control node is at a high level (second control level), which is the opposite of the first control level, the PMOS pull-up transistor Q2 is turned off and the NMOS pull-down transistor is turned on. The output port Lout is connected to the ground terminal GND through the NMOS pull-down transistor, and the output port Lout outputs a low level, turning off the lower bridge power device.
[0027] Since the lower bridge drive circuit 11 in this embodiment is composed of a push-pull structure consisting of a PMOS pull-up transistor Q2 and an NMOS pull-down transistor, rather than a single pull-up transistor forming the output stage, the output port Lout is no longer in a floating state during the turn-off process. Instead, it is clamped to ground GND by the conducting NMOS pull-down transistor, thus providing a low-impedance charge discharge circuit for the gate of the lower bridge power device. This allows the charge accumulated on the gate of the lower bridge power device to be quickly discharged after the turn-off command arrives, and its gate voltage drops rapidly and definitively below the turn-off threshold. This avoids slow or incomplete turn-off due to the lack of discharge for the gate charge, or false turn-on due to the floating gate potential when the switching node voltage changes rapidly, thereby preventing shoot-through of the upper and lower bridge power devices. Meanwhile, since the turn-on branch provides the current source through the PMOS pull-up transistor Q2 and the turn-off branch provides the discharge path through the NMOS pull-down transistor, the two independently undertake the gate charging path and the gate discharging path, respectively. Therefore, by selecting the device dimensions of the PMOS pull-up transistor Q2 and the NMOS pull-down transistor, the turn-on speed and turn-off speed of the lower bridge power device can be independently adjusted to meet the application requirements of different power device gate capacitances and different switching speeds.
[0028] In this embodiment, the triggering unit of the pre-dead-time adaptive circuit 12 is a second D flip-flop U1B, and the post-dead-time reset circuit 13 includes an inverter U3A, a first D flip-flop U2A, and a delayed release branch. The data terminal of the second D flip-flop U1B is connected to the power supply terminal VCC and remains at a high level. Its asynchronous clear terminal is connected to the output terminal of the first D flip-flop U2A to receive a reset signal. Its inverted output terminal is connected to the push-pull control node via the fifth resistor R5. The fifth resistor R5 serves as the common gate drive resistor for the PMOS pull-up transistor Q2 and the NMOS pull-down transistor. The input port Lin is connected to the input terminal of the inverter U3A. The output terminal of the inverter U3A is connected to the clock terminal of the first D flip-flop U2A. The data terminal of the first D flip-flop U2A is connected to the input port Lin. Its asynchronous clear terminal is connected to the power supply terminal VCC and remains inactive. Its output terminal outputs a reset signal. When the downbridge gate control signal changes from the turn-on command level to the turn-off command level, that is, when the input port Lin changes from high level to low level, the inverter U3A inverts the input port Lin and outputs a rising edge. The first D flip-flop U2A latches the turn-off command level (i.e., low level) input to its data terminal under the trigger of this rising edge, so that the output terminal of the first D flip-flop U2A outputs a low-level reset signal and makes the reset signal valid. This valid reset signal is applied to the asynchronous clear terminal of the second D flip-flop U1B, so that the second D flip-flop U1B is asynchronously cleared, its inverted output terminal flips to a high level and through the fifth resistor R5, makes the push-pull control node a second control level, thereby turning off the PMOS pull-up transistor Q2 and turning on the NMOS pull-down transistor. The output port Lout outputs a turn-off level to turn off the downbridge power device and provides a charge discharge circuit for the gate of the downbridge power device. Since the asynchronous reset terminal of the first D flip-flop U2A is always inactive due to its connection to the power supply VCC, the output state of the first D flip-flop U2A is determined only by the data level latched by the rising edge of its clock, thus avoiding uncertain states of the reset signal under power-on or interference conditions.
[0029] The delayed release branch is connected between the output of the first D flip-flop U2A and its asynchronous set terminal. It is used to delay the preset reset hold time after the first D flip-flop U2A outputs a valid reset signal, so that the first D flip-flop U2A is asynchronously set and its output terminal flips to a high level, thus invalidating the reset signal. This releases the asynchronous clearing of the second D flip-flop U1B, allowing the second D flip-flop U1B to regain the ability to respond to the set trigger edge of its clock terminal. Therefore, during the valid period of the reset signal, the second D flip-flop U1B is continuously asynchronously cleared, and its inverting output terminal stably outputs a high level, ensuring the reliable turn-off of the lower bridge power device. Even if the switching node voltage is occasionally lower than the reference voltage Vref due to ringing, noise, or other reasons, resulting in a trough signal, it will not cause the lower bridge power device to be turned on erroneously. After the reset signal becomes invalid after the preset reset holding time, the second D flip-flop U1B can flip when a set trigger edge appears at its clock terminal, causing its inverting output terminal to output a low level. This low level, along with the fifth resistor R5, sets the push-pull control node to the first control level, thereby turning on the PMOS pull-up transistor Q2 and turning off the NMOS pull-down transistor. The output port Lout outputs the turn-on level to turn on the lower bridge power device. Since the data terminal of the second D flip-flop U1B is connected to the power supply terminal VCC and remains at a high level, it is set once every time a set trigger edge appears at its clock terminal. This allows the lower bridge power device to be turned on earlier in each switching cycle after the upper bridge power device is actually turned off, without waiting for the fixed dead time set by the external half-bridge gate drive circuit.
[0030] In this embodiment, the confirmation unit of the pre-dead-time adaptive circuit 12 includes a delay unit and an AND gate. The input of the delay unit is connected to the output of the SW valley detection circuit 14, and is used to delay the valley signal output by the comparator CMP for a preset confirmation time before outputting it. The first input of the AND gate is connected to the output of the SW valley detection circuit 14 to receive the undelayed valley signal, and the second input of the AND gate is connected to the output of the delay unit to receive the valley signal after the preset confirmation time. The output of the AND gate is connected to the clock terminal of the trigger unit, i.e., the second D flip-flop U1B. Thus, only when the voltage of the switching node is continuously lower than the reference voltage Vref, so that the valley signal remains valid within the preset confirmation time, are the undelayed valley signal and the delayed valley signal valid simultaneously. The output of the AND gate flips from an invalid level to an effective level, forming a rising edge to trigger the clock signal of the second D flip-flop U1B. This sets the second D flip-flop U1B, and its inverting output goes low to turn on the lower bridge power device. Conversely, if the voltage of the switching node is only briefly lower than the reference voltage Vref due to ringing, oscillation, or noise, the valley signal disappears within a preset confirmation time. The two inputs of the AND gate cannot be active at the same time, and the output of the AND gate does not generate a trigger edge. The second D flip-flop U1B remains in its original state, and the lower bridge power device is not turned on. This, together with the filtering effect of the filter capacitor C2 and the amplitude discrimination effect of the reference voltage Vref, forms a third barrier against false triggering, preventing the upper and lower bridge power devices from being shot-through.
[0031] In this embodiment, the pre-dead-time adaptive circuit 12 further includes an OR gate; the inverted output of the second D flip-flop U1B is connected to the first input of the OR gate, the second input of the OR gate is connected to the output of the inverter U3A, and the output of the OR gate is connected to the push-pull control node via the fifth resistor R5. Since the output of the inverter U3A is the inverted signal of the input port Lin, when the downbridge gate control signal is the turn-on command level, i.e., the input port Lin is high, the output of the inverter U3A is low. At this time, as long as the inverted output of the second D flip-flop U1B is also low, both inputs of the OR gate are simultaneously low. The output of the OR gate outputs a low level and, via the fifth resistor R5, makes the push-pull control node the first control level, thereby turning on the PMOS pull-up transistor Q2 and turning off the NMOS pull-down transistor, and outputting a low level from the output port Lo. The output ut is turned on to turn on the lower bridge power device; conversely, when the lower bridge gate control signal changes from the turn-on command level to the turn-off command level, that is, when the input port Lin changes from high level to low level, the output of inverter U3A flips to high level, the output of OR gate flips to high level and makes the push-pull control node the second control level, which turns off PMOS pull-up transistor Q2 and turns on NMOS pull-down transistor. The output port Lout outputs the turn-off level to turn off the lower bridge power device and provides a charge discharge circuit for the gate of the lower bridge power device. Thus, the OR gate, connected in parallel with the adaptive turn-on path, forms a bypass turn-on path directly controlled by the lower bridge gate control signal. During the period when the input port Lin is at the turn-on command level, even if the SW valley detection circuit 14 fails to output a valley signal for some reason, or the confirmation unit fails to generate a set trigger edge for some reason, causing the second D flip-flop U1B to fail to be set, the lower bridge power device can still be driven to turn on by this bypass path during the duration of the turn-on command level. At the same time, since either of the two input terminals of the OR gate is high enough to make the push-pull control node the second control level, the turn-off command and the reset signal always logically take precedence over the turn-on command. The turn-off of the lower bridge power device is not affected by this bypass path, thereby ensuring that the upper and lower bridge power devices do not shoot-through while retaining the technical contribution of adaptively shortening the dead time.
[0032] In this embodiment, the SW valley detection circuit 14 further includes a falling edge detection circuit. The input of the falling edge detection circuit is connected to the sampling port SW, and its output is connected to the enable terminal of the comparator CMP. It is used to enable the comparator CMP after a falling edge occurs at the switching node, so that the comparator CMP outputs a valley signal only after a falling edge occurs at the switching node. Specifically, the falling edge detection circuit performs edge detection on the switching node voltage introduced through the current limiting resistor Rsw and the clamping voltage regulation branch. When a falling edge is generated by the switching node voltage transitioning from a high level to a low level, it outputs a valid enable signal to the enable terminal of the comparator CMP from its output terminal. Before this, the comparator CMP is in an disabled state, and its output terminal remains at an invalid level and does not flip with the change of the voltage at the first input terminal. Since the event of the upper bridge power device turning off and the lower bridge power device entering freewheeling in a power half-bridge must manifest as a falling edge from high to low level on the switching node voltage, while the potential drift caused by bus voltage fluctuations, load disturbances, etc., does not have this edge characteristic, by setting a falling edge detection circuit, the comparator CMP only operates during the period after the falling edge occurs at the switching node. This can distinguish the real upper bridge turn-off freewheeling event from abnormal potential disturbances in terms of timing. Thus, the comparator CMP only compares its first input voltage with the reference voltage Vref and outputs a valley signal after the falling edge occurs at the switching node. This, in conjunction with the amplitude discrimination of the reference voltage Vref, the filtering discrimination of the filter capacitor C2, and the duration discrimination of the confirmation unit, further reduces the risk of the comparator CMP erroneously outputting a valley signal due to ringing, oscillation, noise, or other abnormal conditions, which could lead to the lower bridge power device being erroneously turned on, thus avoiding shoot-through of the upper and lower bridge power devices.
[0033] In this embodiment, the half-bridge adaptive dead-time control circuit based on switching node voltage detection of the present invention can be applied to the power half-bridge in a synchronous rectifier converter, synchronous freewheeling converter, or motor drive inverter, where the lower bridge power device is a MOSFET, IGBT, or silicon carbide MOSFET. In the above application, when the upper bridge power device is turned off, since the inductive load current cannot change abruptly, the body diode of the lower bridge power device conducts freewheeling, causing the switching node voltage of the power half-bridge to drop below the reference voltage Vref. The SW valley detection circuit 14 outputs a valley signal accordingly. The pre-dead-time adaptive circuit 12 generates a set trigger edge after the valley signal remains valid for a preset confirmation time, causing the output of the trigger unit, i.e., the second D flip-flop U1B, to set the push-pull control node to the first control level, thereby turning on the lower bridge power device before the fixed dead time set by the external half-bridge gate drive circuit ends. That is, the moment when the output of the trigger unit sets the push-pull control node to the first control level is after the upper bridge power device is turned off and before the built-in dead time of the external half-bridge gate drive circuit ends. Therefore, on the one hand, since the lower bridge power device is turned on only after the upper bridge power device has actually turned off and the switching node voltage has dropped below the reference voltage Vref, shoot-through between the upper and lower bridge power devices will not occur, ensuring the safe operation of the power half-bridge. On the other hand, since the turn-on of the lower bridge power device is no longer constrained by the fixed dead time preset by the external half-bridge gate drive circuit, but is turned on immediately after the upper bridge power device has actually turned off, the freewheeling time of the body diode of the lower bridge power device can be significantly shortened, reducing freewheeling losses and device heating, improving conversion efficiency, and suppressing switching node waveform oscillations and electromagnetic interference caused by excessive dead time. Since the voltage drop time of the aforementioned switching node is determined by the actual switching speed of the power device, the magnitude of the load current, and the operating conditions, the dead time in this embodiment can adaptively change with different power device types, different switching speeds, different loads, and different operating conditions without prior testing and writing correction parameters, and without the need for high-speed analog-to-digital sampling and current detection by the microprocessor. Therefore, it can be widely used in power supply and motor drive fields containing half-bridge topology components, such as half-bridge and full-bridge.
[0034] In this embodiment, the lower bridge drive circuit 11, the SW valley detection circuit 14, the pre-dead-zone adaptive circuit 12, and the post-dead-zone reset circuit 13 are integrated into a single chip. The input port Lin, the output port Lout, the sampling port SW, the power supply terminal VCC, and the ground terminal GND are the five pins of this single chip, thus making the half-bridge adaptive dead-zone control circuit based on switching node voltage detection of the present invention a five-pin independent chip product. When in use, this chip is connected in series between the lower bridge drive output of the external half-bridge gate drive circuit and the gate of the lower bridge power device of the power half-bridge. Its input port Lin is connected to the lower bridge drive output of the external half-bridge gate drive circuit to receive the lower bridge gate control signal. Its output port Lout is connected to the gate of the lower bridge power device to output the turn-on or turn-off level. Its sampling port SW is connected to the switching node of the power half-bridge to sample the switching node voltage. Its power supply terminal VCC and ground terminal GND share the same power rail with the external half-bridge gate drive circuit, that is, it is powered by the same power supply of the external half-bridge gate drive circuit. There is no need to configure an independent power supply circuit for this chip, thereby simplifying the power supply design and peripheral wiring of the application. The power supply terminal VCC serves two purposes: firstly, it powers the internal components of the circuit, including the comparator CMP, inverter U3A, first D flip-flop U2A, and second D flip-flop U1B; secondly, it provides a current source for the gate of the lower-bridge power device. Specifically, when the PMOS pull-up transistor Q2 is turned on, current is output through the output port Lout to charge the gate of the lower-bridge power device, thus turning it on. Since this embodiment integrates the above circuitry onto a single chip with five pins, the application only needs to connect this chip in series with the lower-bridge drive path of an existing half-bridge gate drive circuit, and connect an external current-limiting resistor Rsw and a gate drive resistor. This allows for the addition of adaptive dead-time control functionality to existing half-bridge power supplies or motor drive systems without modifying the existing drive chip, adding a microprocessor, or performing prior parameter calibration. Therefore, it is widely applicable to various power supply and motor drive applications containing half-bridge topologies, such as synchronous rectifier converters, synchronous freewheeling converters, and motor drive inverters.
[0035] Compared with the prior art, the embodiments of this application have the following beneficial effects: First, by directly detecting the actual voltage state of the power half-bridge switching nodes, the actual turn-off time of the upper-bridge power devices is determined, allowing the dead time to automatically adapt to the power device type, switching speed, load size, and operating conditions, eliminating the need to wait for the fixed dead time set by the external half-bridge gate drive circuit. This significantly shortens the freewheeling time of the lower-bridge power device's body diode, reduces freewheeling losses, and improves conversion efficiency.
[0036] Secondly, since the dead time no longer causes the body diode to continue flowing for an extended period due to an excessively large value, nor causes the upper and lower bridge power devices to shoot through due to an excessively small value, this solution fundamentally eliminates problems such as overheating and device failure caused by improper dead time. Simultaneously, the pull-down switch in the lower bridge drive circuit provides a low-impedance charge discharge path for the gate of the lower bridge power device, allowing the gate voltage to drop rapidly and definitively below the turn-off threshold, avoiding incomplete turn-off or false turn-on due to gate floating.
[0037] Third, a four-pronged mechanism—reference voltage amplitude discrimination, filter capacitor discrimination, confirmation unit duration discrimination, and post-dead-zone reset circuit reset duration blocking—is used to eliminate false trough signals caused by abnormal conditions such as ringing, oscillation, and noise. Therefore, even if the switching node voltage is accidentally lower than the reference voltage due to interference, the lower bridge power device will not be mistakenly turned on, fundamentally preventing shoot-through between the upper and lower bridge power devices.
[0038] Fourth, the dead time is determined in real time based on the switching node voltage, eliminating the need for proactive testing of dead times under different power devices, switching speeds, loads, temperatures, and operating conditions beforehand and pre-programming for calibration. This eliminates the tedious preliminary testing and parameter calibration work, and avoids limitations and calibration errors caused by limited calibration samples.
[0039] Fifth, adaptive dead-time control is achieved using simple hardware circuits such as comparators, trigger units, and logic gates, eliminating the need for a microprocessor for high-speed analog-to-digital conversion, voltage sampling, zero-crossing detection, and current detection. Therefore, the circuit structure is simple, small in size, low in cost, and technically easy to implement, making it widely applicable to various application markets.
[0040] Sixth, it is compatible with various power devices such as MOSFETs, IGBTs, and silicon carbide MOSFETs, and can be applied to applications containing half-bridge topologies, such as synchronous rectifier converters, synchronous freewheeling converters, and motor drive inverters. Furthermore, by selecting the tiered value of the current-limiting resistor Rsw in the input branch, it can adapt to different bus voltage levels, making it suitable for both low-voltage and high-voltage applications.
[0041] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.
Claims
1. A half-bridge adaptive dead-time control circuit based on switching node voltage detection, characterized in that, The lower bridge drive output terminal of the external half-bridge gate drive circuit and the gate of the lower bridge power device of the power half-bridge are connected in series, including the input port Lin, the output port Lout, the sampling port SW, the power supply terminal VCC and the ground terminal GND, as well as the lower bridge drive circuit (11), the SW valley detection circuit (14), the pre-dead-zone adaptive circuit (12) and the post-dead-zone reset circuit (13). The input port Lin is used to connect to the lower bridge drive output of the external half-bridge gate drive circuit and receive the lower bridge gate control signal. The lower bridge gate control signal has an on command level and an off command level. The lower bridge drive circuit (11) includes a pull-up switch, a pull-down switch, and a push-pull control node; the pull-up switch is connected between the power supply terminal VCC and the output port Lout, the pull-down switch is connected between the output port Lout and the ground terminal GND, and the control terminal of the pull-up switch and the control terminal of the pull-down switch are connected to form a push-pull control node; the output port Lout is used to connect the gate of the lower bridge power device. When the push-pull control node is at the first control level, the pull-up switch is turned on and the pull-down switch is turned off, and the output port Lout outputs an on level to turn on the lower bridge power device; when the push-pull control node is at the second control level, which is the opposite of the first control level, the pull-up switch is turned off and the pull-down switch is turned on, and the output port Lout outputs a off level to turn off the lower bridge power device and provide a charge discharge circuit for the gate of the lower bridge power device. The SW valley detection circuit (14) includes a comparator. The first input terminal of the comparator is connected to the sampling port SW via the input branch. The sampling port SW is used to connect to the switching node of the power half-bridge. The second input terminal of the comparator is connected to the reference voltage. When the voltage of the switching node is lower than the reference voltage, the output terminal of the comparator flips from an invalid level to an effective level and outputs a valley signal. The input terminal of the post-dead-time reset circuit (13) is connected to the input port Lin, which is used to generate a valid reset signal when the lower bridge gate control signal changes from the turn-on command level to the turn-off command level, and the reset signal becomes invalid after the turn-off command level appears and continues to be preset for a reset holding time. The pre-dead-zone adaptive circuit (12) includes a trigger unit and a confirmation unit; the set input terminal of the trigger unit is connected to a fixed valid level, the asynchronous reset terminal of the trigger unit is connected to the output terminal of the post-dead-zone reset circuit (13) to receive a reset signal, and the output terminal of the trigger unit is connected to a push-pull control node; the confirmation unit is connected between the output terminal of the SW valley detection circuit (14) and the clock terminal of the trigger unit, and is used to output a set trigger edge to the clock terminal of the trigger unit after the valley signal has been valid for a preset confirmation time. When the reset signal is invalid and a set trigger edge appears at the clock terminal of the trigger unit, the output terminal of the trigger unit sets the push-pull control node to the first control level to turn on the lower bridge power device; when the reset signal is valid, the trigger unit is asynchronously reset, and the output terminal of the trigger unit sets the push-pull control node to the second control level to turn off the lower bridge power device.
2. The half-bridge adaptive dead-time control circuit as described in claim 1, characterized in that, The input branch includes a current-limiting resistor and a clamping voltage regulator branch. The SW valley detection circuit (14) also includes a filter capacitor. The sampling port SW is connected to the first end of the clamping voltage regulator branch via the current-limiting resistor. The second end of the clamping voltage regulator branch is connected to the ground GND. The first end of the clamping voltage regulator branch is connected to the first input end of the comparator. The filter capacitor is connected between the first input end of the comparator and the ground GND. The clamping voltage regulator branch includes a fourth resistor and a first Zener diode. The first end of the fourth resistor is connected to the end of the current-limiting resistor furthest from the sampling port SW and serves as the first end of the clamping voltage regulator branch. The second end of the fourth resistor serves as the second end of the clamping voltage regulator branch and is connected to ground GND. The cathode of the first Zener diode is connected to the first end of the clamping voltage regulator branch, and the anode of the first Zener diode is connected to ground GND. The resistance value of the current-limiting resistor is selected according to the bus voltage level of the power half-bridge, so that the operating current of the first Zener diode when the voltage at the switching node is the bus voltage is between the minimum stable breakdown current and the maximum allowable breakdown current of the first Zener diode.
3. The half-bridge adaptive dead-time control circuit as described in claim 2, characterized in that, The filter capacitor is used to filter the ringing voltage of the switching node, so that the voltage at the first input terminal of the comparator is lower than the reference voltage only when the upper bridge power device of the power half-bridge is completely turned off and the lower bridge power device is freewheeling.
4. The half-bridge adaptive dead-time control circuit as described in claim 1, characterized in that, The power-on command level is high, and the power-off command level is low; the power-on level is high, and the power-off level is low. The pull-up switch is a PMOS pull-up transistor (Q2), and the pull-down switch is an NMOS pull-down transistor. The source of the PMOS pull-up transistor (Q2) is connected to the power supply terminal VCC, and the drain is connected to the output port Lout. The source of the NMOS pull-down transistor is connected to the ground terminal GND, and the drain is connected to the output port Lout. The gates of the PMOS pull-up transistor (Q2) and the NMOS pull-down transistor are connected to form a push-pull control node. The first control level is low, and the second control level is high. The output port Lout is connected to the gate of the lower bridge power device through a gate drive resistor.
5. The half-bridge adaptive dead-time control circuit as described in claim 1, characterized in that, The triggering unit is a second D flip-flop, and the post-dead-zone reset circuit (13) includes an inverter, a first D flip-flop, and a delayed release branch; The data terminal of the second D flip-flop is connected to the power supply VCC and kept at a high level. The asynchronous clear terminal of the second D flip-flop is connected to the output terminal of the first D flip-flop. The inverted output terminal of the second D flip-flop is connected to the push-pull control node via the fifth resistor. The input port Lin is connected to the input of the inverter, the output of the inverter is connected to the clock terminal of the first D flip-flop, the data terminal of the first D flip-flop is connected to the input port Lin, the asynchronous clear terminal of the first D flip-flop is connected to the power supply VCC, and the output of the first D flip-flop outputs a reset signal. When the lower bridge gate control signal changes from the turn-on command level to the turn-off command level, the inverter outputs a rising edge, the first D flip-flop latches the turn-off command level and makes the reset signal valid; The delayed release branch is connected between the output of the first D flip-flop and its asynchronous set terminal. It is used to delay the preset reset hold time after the first D flip-flop outputs a valid reset signal, so that the first D flip-flop is asynchronously set and the reset signal becomes invalid.
6. The half-bridge adaptive dead-time control circuit as described in claim 1, characterized in that, The confirmation unit includes a delay unit and an AND gate; the input of the delay unit is connected to the output of the SW valley detection circuit (14) and is used to delay the valley signal for a preset confirmation time; the first input of the AND gate is connected to the output of the SW valley detection circuit (14), the second input of the AND gate is connected to the output of the delay unit, and the output of the AND gate is connected to the clock terminal of the trigger unit.
7. The half-bridge adaptive dead-time control circuit as described in claim 5, characterized in that, The pre-dead-zone adaptive circuit (12) also includes an OR gate; the inverted output of the second D flip-flop is connected to the first input of the OR gate, the second input of the OR gate is connected to the output of the inverter, and the fifth resistor of the output of the OR gate is connected to the push-pull control node. When the lower bridge gate control signal is at the turn-on command level, the output of the OR gate sets the push-pull control node to the first control level, thus turning on the pull-up switch.
8. The half-bridge adaptive dead-time control circuit as described in claim 1, characterized in that, The SW valley detection circuit (14) also includes a falling edge detection circuit; the input of the falling edge detection circuit is connected to the sampling port SW, and the output of the falling edge detection circuit is connected to the enable terminal of the comparator, which is used to enable the comparator after the falling edge appears at the switching node, so that the comparator outputs the valley signal only after the falling edge appears at the switching node.
9. The half-bridge adaptive dead-time control circuit as described in claim 1, characterized in that, The power half-bridge is a synchronous rectifier converter, synchronous freewheeling converter, or motor drive inverter half-bridge; the lower bridge power device is a MOSFET, IGBT, or silicon carbide MOSFET. The body diode of the lower bridge power device conducts freewheeling after the upper bridge power device is turned off, causing the voltage of the switching node to drop below the reference voltage; the moment when the output of the trigger unit sets the push-pull control node to the first control level is after the upper bridge power device is turned off and before the end of the built-in dead time of the external half-bridge gate drive circuit.
10. The half-bridge adaptive dead-time control circuit as described in any one of claims 1 to 9, characterized in that, The lower bridge drive circuit (11), SW valley detection circuit (14), pre-dead-zone adaptive circuit (12), and post-dead-zone reset circuit (13) are integrated into a single chip. The input port Lin, output port Lout, sampling port SW, power supply VCC, and ground GND are the five pins of the single chip. The power supply VCC and ground GND share the same power rail with the external half-bridge gate drive circuit. The power supply VCC is used to power the half-bridge adaptive dead-zone control circuit based on the switching node voltage detection and to provide a current source for the gate of the lower bridge power device.
