An adaptive rectifier driving regulation method and system for trench-type schottky chip

Abstract

The application discloses an adaptive rectification driving regulation method and system suitable for a trench type Schottky chip, and comprises the following steps: S1, parameter pre-marking; S2, sampling and separating a leakage current; S3, backstepping a junction temperature; S4, extracting an equivalent junction capacitance; S5, calculating a turn-off slope and regulating; S6, calculating a gate bias voltage; S7, applying an adaptive gate voltage; and S8, bidirectional closed-loop linkage. According to the application, the real-time junction temperature of the chip is backstepped through a temperature drift model, an expensive on-chip temperature sensor is saved, and the defect attribute (leakage temperature drift characteristic) of the device is converted into accurate working condition sensing resources; and through the corrected algorithm, the gate oxide thickness is introduced to complete dimension correction, the turn-off slope is automatically limited under high-temperature or large-current working conditions, the electric field concentration and voltage ringing of the trench corner are effectively inhibited, the gate oxide dielectric aging is greatly delayed, and the service life of the device is prolonged.

Description

Technical Field

[0001] This invention relates to the field of power semiconductor device drive control technology, specifically to an adaptive rectification drive control method and system adapted to trench-type Schottky chips. Background Technology

[0002] Trench MOS Barrier Schottky (TMBS) diodes utilize a depletion layer on the trench sidewalls to pinch off the surface electric field by etching a deep trench below the anode, oxidizing the sidewalls, and filling it with a polysilicon gate. This allows the TMBS diode to maintain a low forward voltage drop. While offering advantages, it significantly improves the reverse breakdown voltage of the device.

[0003] However, TMBS is a unipolar device with a reverse leakage current. junction temperature Extremely sensitive (following an exponential law). Under high temperature and high back pressure conditions, The rapid increase leads to a further rise in junction temperature, which can easily trigger irreversible thermal breakdown. Existing technologies typically rely on external NTC thermistors for coarse temperature protection, which are slow to respond and cannot detect the true hot spot temperature inside the chip; furthermore, the deep trench structure of TMBS introduces complex parasitic capacitances (especially gate-anode capacitance). ), and the parasitic inductance of the power circuit LC oscillations (ringing) occur at the moment of high-frequency turn-off. Severe ringing not only causes EMI problems, but also creates a local electric field at the corner of the trench that is much higher than the average value, accelerating the aging of the gate oxide dielectric; while the parameters of the existing RC absorption circuit are fixed and cannot be dynamically adjusted according to the operating conditions. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an adaptive rectification drive control method and system adapted to trench-type Schottky chips, solving the problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an adaptive rectification drive control method for trench-type Schottky chips, wherein the trench-type Schottky chip is a trench-type MOS barrier Schottky diode with an independently led-out gate, and its trench gate is electrically connected to an external drive pin through a passivation layer opening; comprising the following steps: Step S1, Parameter pre-calibration: During the chip power-on initialization phase, the leakage current temperature sensitivity coefficient is obtained through temperature control testing. and voltage sensitivity And it is stored in non-volatile memory; Step S2: Within the turn-off steady-state window of each switching cycle, sample the anode voltage of the trench Schottky chip. With anode current From the anode current The reverse leakage current at the current moment is obtained by separation. ; Step S3, based on the reverse leakage current With current reverse bias Substituting the pre-stored leakage current temperature drift model, the junction temperature of the chip can be calculated. ; Step S4, the rate of change of voltage during the turn-off transient. In this stage, the equivalent junction capacitance under the current bias voltage is extracted online based on the displacement current formula. ; Step S5, set the junction temperature Reverse leakage current Equivalent junction capacitance Reverse bias Pre-stored parasitic inductance of power circuit and gate oxide thickness Substitute into the adaptive turn-off slope formula to calculate the target turn-off slope. and will The output is sent to the drive edge control module to regulate the shutdown process. ; Step S6, read the current load current Ringing amplitude in this cycle Combined with the results obtained in step S3 Substituting the result of step S5 into the gate bias adaptive formula, the trench gate bias voltage is calculated. ; Step S7, the... The output is sent to the trench gate drive stage, and during the forward conduction period, it is based on... Apply grid pressure to cause the trench sidewalls to form inverted auxiliary channels or deepen the depletion layer clamping as needed; Step S8, set the junction temperature for this cycle. Ringing amplitude and reverse leakage current Write the parameters into the linkage register as input parameters for steps S5 and S6 in the next cycle, forming a two-way closed-loop linkage control of the turn-off slope and gate bias.

[0006] Furthermore, the method for defining the turn-off steady-state window in step S2 is as follows: monitoring the differential of the anode voltage after the turn-off edge ends. ,when hour, The maximum slope during the initial turn-off phase is defined as a time window lasting 50 ns to 200 ns; within this window, the anode current... The displacement current component in the middle approaches zero, therefore The reverse bias voltage Take the contents of this window The arithmetic mean.

[0007] Furthermore, the leakage current temperature drift model mentioned in step S3 is as follows:

[0008] in, Reference leakage current; For reference junction temperature, the value is taken as follows. , The reference value under the rated reverse test voltage is determined by the temperature control test in step S1 and is used in step S1. Fit calibration; It is the base of the natural logarithm; The leakage current temperature sensitivity coefficient has a range of values. ; The leakage current voltage sensitivity is derived from the temperature control test fitting in step S1; The junction temperature, which is the unknown, can be obtained by transforming and inversely solving the above formula:

[0010] Furthermore, the adaptive turn-off slope formula described in step S5 is:

[0011] in The absolute turn-off slope is expressed in V / ns. , which is the withstand voltage of the dielectric layer at the trench corner, in volts (V). The critical breakdown electric field of the dielectric layer at the trench corner. The physical thickness of the trench gate oxide dielectric layer is used to convert the electric field strength into voltage. The reference voltage is the reverse bias voltage corresponding to the device's rated operating voltage. The value, in V, is used for dimensional normalization. The rated operating voltage of the device is the maximum reverse operating voltage specified in the chip datasheet; for example, for a TMBS device with a rated operating voltage of 100V. Take the device under a 100V reverse bias voltage. The values; devices with different rated operating voltages are determined according to the same principle. value; This is the total parasitic inductance of the power circuit; The transient peak current before the current cycle is turned off is derived from the anode Shunt resistor sampling circuit; This refers to the temperature drift-slope mapping coefficient; 25℃ The reference leakage current under the given conditions is derived from the chip datasheet. The native shutdown slope of the driver chip under rated operating conditions is derived from the driver IC datasheet. The critical breakdown electric field of the dielectric layer at the trench corner was obtained by TCAD device simulation. The simulation adopted a drift-diffusion model with a minimum mesh size of 5nm and Neumann boundary conditions on the trench sidewall. 80% of the breakdown electric field obtained from the simulation was taken as the derating value. : Taken from chip manufacturing process design rules; : Obtained through a pre-characterization database of parasitic parameters of PCB layout and package structure; The value ranges from 0.5 to 2, and is determined by system loop stability simulation.

[0012] Furthermore, the online extraction described in step S4 The specific numerical calculation method is as follows:

[0013] The integration interval is selected as 10% to 90% of the envelope of the turn-off transient voltage, i.e. , In the formula Used to eliminate the DC component of leakage current and retain the AC displacement current used for charging and discharging junction capacitance.

[0014] Furthermore, the gate offset adaptive formula described in step S6 is:

[0015] in, It serves as the zero bias or slight negative voltage reference for the trench gate, and its value is derived from the rated setting of the TMBS device structure. This is the current-gate gain coefficient, in V / A, calibrated by TCAD simulation, reflecting the increase in gate voltage and decrease in forward voltage drop. Marginal revenue; The surge-assisted trigger current threshold is set to 80% of the chip's rated current, derived from the device... The derating settings in the specification sheet; for The smoothing scaling factor of the function is used to prevent logic jitter and comes from the system's anti-jitter design. The leakage current penalty factor is the maximum allowable leakage current of the system. By reverse deduction, Derived from the system thermal design specifications; The ringing-loss compensation coefficient is determined by the overall system efficiency trade-off. The ringing amplitude detected in this cycle is derived from the peak sampling of the waveform turned off in step S5; The upper limit of the ringing amplitude allowed by the system is derived from EMI and device withstand voltage design specifications; Typical values ​​for each parameter in the above formula: =0V; The value range is 2~3 V / A; Set to 80% of the chip's rated current; The value range is 10% to 20% of the rated current, and its specific value is determined based on the statistical distribution of the load current fluctuation amplitude: collect the load current fluctuation data under normal operating conditions of the system, calculate its standard deviation, and take... The value is two standard deviations and is limited to 10% to 20% of the rated current; if no measured data is available, 15% of the rated current is used by default. Depend on Reverse reasoning, Derived from the system thermal design specifications; the inverse relationship is: ,in This is the maximum allowable positive bias voltage for the gate. The reference gate voltage; this inverse relationship ensures that when the leakage current... Reaching 0.8 At that time, leakage current penalty term Sufficient to offset surge auxiliary gain term The maximum value, ensuring To achieve safety protection under high-temperature operating conditions; in specific numerical terms, for example when , hour, However, this value needs to be adjusted according to the overall system gain distribution, and its general range is [range missing]. In the specific embodiments of this application, the following are taken: ,correspond Design margin scenario; if change, Adjust accordingly based on the above relationships; The value range is 1~2 V.

[0016] Furthermore, step S6 also includes security protection logic: when a detection is detected... At that time, the system will be forcibly locked. Any form of gate lifting operation is prohibited to prevent thermal runaway caused by leakage current at high temperatures.

[0017] Furthermore, the bidirectional closed-loop linkage mechanism described in step S8 specifically includes: Forward link: Output of step S3 and the output of step S5 Stored in the linkage register for calculation in step S6. The basis for the judgment at that time; Reverse link: Output of step S7 The width of the depletion region on the trench sidewall was altered, resulting in the extraction in the next cycle step S4. The value is offset, and this offset is used to monitor the aging status of the device or for advanced prediction algorithms.

[0018] Furthermore, the trench gate bias described in step S7 The scope of implementation is clamped to or Between; where positive voltage is used to form an inverse channel to reduce the forward voltage drop, and negative voltage is used to deepen depletion to suppress high-temperature leakage current; and 10ns before the turn-off signal is issued, the gate voltage is forcibly reset to or This eliminates turn-off noise introduced by gate-anode capacitive coupling; at the same time, during the conduction period, it ensures that the difference between the anode potential and the gate potential does not exceed the gate oxide tolerance limit.

[0019] An adaptive rectification drive control system adapted to trench Schottky chips, the system applying the aforementioned adaptive rectification drive control method adapted to trench Schottky chips, comprising: The parameter pre-calibration module is used to obtain the leakage current temperature sensitivity coefficient through temperature control testing during the chip power-on initialization phase. and voltage sensitivity And it is stored in non-volatile memory; The leakage current sampling module is used to sample the anode voltage of the trench Schottky chip within the turn-off steady-state window of each switching cycle. With anode current From the anode current The reverse leakage current at the current moment is obtained by separation. ; Junction temperature reverse calculation module, used to calculate based on the reverse leakage current With current reverse bias Substituting the pre-stored leakage current temperature drift model, the junction temperature of the chip can be calculated. ; Junction capacitance extraction module for voltage change rate during turn-off transients. In this stage, the equivalent junction capacitance under the current bias voltage is extracted online based on the displacement current formula. ; The turn-off slope calculation module is used to calculate the junction temperature. Reverse leakage current Equivalent junction capacitance Reverse bias Pre-stored parasitic inductance of power circuit and gate oxide thickness Substitute into the adaptive turn-off slope formula to calculate the target turn-off slope. and will The output is sent to the drive edge control module to regulate the shutdown process. ; The grid bias calculation module is used to read the current load current. Ringing amplitude in this cycle The junction temperature back-calculation module provides the following results: The results from the turn-off slope calculation module are substituted into the gate bias adaptive formula to calculate the trench gate bias voltage. ; Gate drive module, used to drive the gate drive module. The output is sent to the trench gate drive stage, and during the forward conduction period, it is based on... Apply grid pressure to cause the trench sidewalls to form inverted auxiliary channels or deepen the depletion layer clamping as needed; The closed-loop linkage register is used to set the junction temperature for the current cycle. Ringing amplitude and reverse leakage current The parameters are written into the linkage register and used as input parameters for the turn-off slope calculation module and the gate bias calculation module in the next cycle, forming a two-way closed-loop linkage control of the turn-off slope and the gate bias.

[0020] This invention provides an adaptive rectification drive control method and system for trench-type Schottky chips, which has the following advantages: 1. This adaptive rectification drive control method and system for trench-type Schottky chips utilizes the inherent characteristics of TMBS. As a temperature probe, the real-time junction temperature of the chip can be inferred by using a temperature drift model. This eliminates the need for expensive on-chip temperature sensors, transforming the device's "defect attributes" (leakage and temperature drift characteristics) into precise operational condition sensing resources; and through the corrected... Algorithm, introducing gate oxide thickness Dimensional correction was completed, and the critical electric field and parasitic inductance of the coupling medium were integrated. It also features leakage current and temperature drift characteristics. Under high temperature or high current conditions, it automatically limits the turn-off slope, effectively suppresses electric field concentration and voltage ringing at trench corners, significantly delays gate oxide aging, and extends device lifespan.

[0021] 2. This adaptive rectification drive control method and system for trench-type Schottky chips utilizes multi-dimensional coupling... An adaptive algorithm automatically raises the gate voltage under heavy load and surge conditions, and utilizes the inversion layer on the trench sidewall to assist conduction, significantly reducing the device's forward voltage drop. Applying negative pressure grid offset under high temperature and light load conditions deepens the depletion layer on the trench sidewalls, effectively suppressing... The surge eliminates the risk of thermal runaway from a mechanistic perspective; it also constructs a two-way closed-loop linkage mechanism of "turn-off stress-conduction loss". When the turn-off speed is reduced and the switching loss is increased in order to suppress EMI ringing, the system automatically fine-tunes the gate voltage to compensate for the loss, breaking through the limitations of traditional single parameter optimization and achieving multi-objective global optimization of system efficiency, device reliability and EMI performance. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the steps of an adaptive rectification drive control method and system for trench-type Schottky chips according to the present invention. Detailed Implementation

[0023] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0024] like Figure 1 As shown, this invention provides a technical solution: an adaptive rectification drive control method and system adapted to trench-type Schottky chips. The system applies an adaptive rectification drive control method adapted to trench-type Schottky chips. A specific embodiment of this method, combined with a 100V / 40A gate-independent lead-out TMBS (GCSD structure) applied to a 48V to 12V synchronous rectification Buck circuit, is provided to illustrate the technical solution of this invention in detail and completely, specifically including: Chip structure: 100V-level TMBS device with independent gate lead-out, trench depth 3μm, gate oxide dielectric thickness... The device trench gate is led out to an independent pad through an opening in the chip passivation layer, achieving complete electrical isolation from the cathode, which is different from the traditional TMBS structure with a short gate and cathode.

[0025] Pre-calibration step (corresponding to step S1): During the chip power-on initialization phase, a constant temperature calibration platform is built, and the chip junction temperature is controlled to [temperature value missing]. Apply rated reverse test voltage Collect the corresponding reverse leakage current point by point. Using the least squares method With junction temperature Linear fitting was performed to obtain the leakage current temperature sensitivity coefficient. The leakage current-voltage sensitivity was obtained by fitting multiple sets of reverse bias voltage scanning tests. All calibration parameters are stored in the driver chip's non-volatile memory for use by real-time algorithms.

[0026] Sampling and driving hardware: A 5mΩ high-precision shunt resistor is connected in series with the anode of the device, and a 100MHz differential amplifier is connected to ADC1 to realize the anode current. High-speed sampling; the anode is connected to ADC2 via a 10:1 high-voltage differential probe to realize the anode voltage. Real-time data acquisition. Using an FPGA as the core control unit, coupled with positive and negative voltage charge pumps, it can output... Adjustable gate voltage to meet adaptive gate bias control requirements.

[0027] Step S2: Leakage Current Precise separation During each PWM switching cycle, after the main driver issues a turn-off signal, the FPGA monitors the voltage change rate in real time. Turn-off transient interval For high slope The phase then enters a 100 ns turn-off steady-state window. This window satisfies The displacement current component is approximately zero. The FPGA samples the anode current within the steady-state window to obtain the reverse leakage current for this cycle. The average anode voltage within the window is filtered to obtain the real-time reverse bias voltage. .

[0028] Step S3: Junction Temperature High-precision reverse propulsion Recalling pre-stored calibration parameters: reference temperature Reference leakage current (Typical values ​​from the chip datasheet at 25℃ / 80V), reference reverse voltage Thermosensitive coefficient Voltage sensitivity Substituting into the junction temperature back-calculation formula:

[0029] The original junction temperature was calculated. Through first-order lag filtering (filter coefficients) Smoothing and noise reduction to obtain real-time effective junction temperature , latched into linkage register A.

[0030] Step S4: Equivalent junction capacitance Online extraction During the shutdown transient The DC component of leakage current is eliminated point by point within the interval. Calculate the transient junction capacitance over the voltage range. Internal numerical integral average:

[0031] The equivalent junction capacitance is calculated in this period. It is used for monitoring the aging status of devices and subsequent closed-loop control.

[0032] Step S5: Adaptive Shutdown Slope Calculation and Execution Loading system preset parameters: critical breakdown electric field Gate oxide thickness Therefore, withstand voltage In this embodiment, the rated operating voltage of the device is 100V, corresponding to the reference voltage. For devices with different rated operating voltages, Following the same principle, the rated voltage is taken. The value, for example, the corresponding value for a device with a rated operating voltage of 60V. Devices with a rated operating voltage of 150V are corresponding to And so on. Parasitic inductance Turn-off peak current Temperature drift mapping coefficient Reference leakage current Native shutdown slope .

[0033] Calculate the ringing suppression safety slope term:

[0034] Calculate the adaptive correction slope term for temperature drift:

[0035] The optimal adaptive slope is obtained by minimizing the two values. Configure the current source parameters of the FPGA driver chip to precisely control the shutdown. Simultaneously, the peak value of the ringing during the current cycle is collected. Stored in linkage register B.

[0036] Step S6: Gate bias voltage Multidimensional adaptive decision Loading control parameters: reference gate voltage Current gain coefficient Current threshold , scale factor Leakage current penalty coefficient Maximum permissible leakage current Ringing compensation coefficient Maximum ringing limit Real-time load current .

[0037] Calculate each control component separately: Surge-assisted gain items:

[0038] High-temperature leakage current penalty:

[0039] Ringing loss compensation item:

[0040] The original gate voltage is obtained by superposition: .

[0041] Security logic determination: Current Thermal runaway lockout protection was not triggered; after voltage clamping, the gate voltage eventually adapted. (Limited to the +3V limit, the actual output is +2.04V).

[0042] Step S7: Precise gate drive timing coordination During the next cycle's device turn-on phase, when a forward voltage drop is detected... At that time, the FPGA controls the charge pump output. Adaptive gate voltage forms a weak inverse channel on the trench sidewall to help reduce conduction loss.

[0043] Full-range potential protection: Real-time monitoring of the gate-anode potential difference during the conduction cycle, strictly controlled within the gate oxide tolerance limit ( Within a certain range, avoid dielectric breakdown failure.

[0044] Pre-shutdown reset: 10ns before the shutdown signal is sent, the gate voltage is reset to [value missing]. This ensures that the depletion layer on the trench sidewall is fully fractured, completely eliminating the parasitic capacitance of the grid anode. Coupling-induced shutdown noise and erroneous switching problems.

[0045] Step S8: Two-way closed-loop linkage update After this period's adjustment ends, the junction temperature will be smoothed out. Ringing amplitude The synchronous update is sent to the linkage register, serving as the core input for slope control and gate voltage adaptation in the next cycle. When subsequent operating conditions increase temperature and leakage current... When the voltage increases, the system will automatically reduce the turn-off slope and strengthen the gate voltage negative pressure penalty, forming a two-way closed-loop adaptive control of "temperature-leakage-ringing-gate voltage-slope", achieving a dynamic balance between device safety and system efficiency.

[0046] To verify the effectiveness of the method of the present invention, a traditional fixed drive scheme was set as a comparative example, and comparative experiments were carried out on the same hardware platform (100V / 40A gate independently led-out TMBS, 48V to 12V synchronous rectification Buck circuit) and under the same test conditions. The scaling settings are as follows: Gate drive: TMBS gate directly grounded ( =0V), The potential of the gate to ground (cathode) is that of the trench gate, and there is no adaptive bias function; the turn-off slope is fixed at 10V / ns and does not adjust with the operating conditions; and there is no leakage current temperature drift compensation, no ringing suppression, no junction temperature reverse push, and only passive suppression of oscillation by relying on the traditional RC absorption network. The test conditions are as follows: Ambient temperature: 125 degrees Celsius (high temperature oven); Load condition: full load output ( Output voltage The switching frequency was 200kHz; the testing instruments included a 4-channel 1GHz oscilloscope, an infrared thermal imager, a precision power analyzer, and a surge generator; the test results are as follows: The comparative example exhibits severe LC oscillations at the anode voltage during turn-off, with a ringing peak reaching 25V, superimposed on the 48V bus. The device experiences a voltage stress of up to 73V, approaching the 100V withstand voltage margin limit, and the EMI spectrum exceeds acceptable limits. This invention addresses this issue through adaptive turn-off slope control (…). ) and gate negative voltage reset ( The dual effect of 0.5V suppresses the ringing amplitude to 12V, a reduction of 52%. This not only significantly improves EMI performance but also reduces device voltage stress to below 60V, preserving sufficient safety margin for the trench corner dielectric layer and effectively delaying the gate oxide aging process.

[0047] The comparative example shows leakage current at high temperatures due to a fixed turn-off slope. Rapidly increased (actual measurement) This leads to positive feedback heat accumulation, with the highest junction temperature reaching 118 degrees Celsius, approaching the device's temperature. The limit of this invention is reached by in-situ deducing the junction temperature from the leakage current (S3), automatically reducing the turn-off slope (S5) and applying a negative gate bias penalty (S6) under high-temperature conditions. It fell back to 1.5 The maximum junction temperature was ultimately controlled at 103 degrees Celsius, a reduction of 15 degrees Celsius compared to the control, completely eliminating the risk of thermal runaway and fundamentally ensuring the long-term reliability of the device.

[0048] The comparative example has a full-load efficiency of 92.1%, with the main losses stemming from switching losses and ringing losses caused by a fixed turn-off slope. The present invention addresses this by using a positive gate voltage (+2.04V) to assist turn-on under heavy load, thereby reducing losses. Meanwhile, the adaptive turn-off slope maximizes the switching speed while ensuring media safety, thereby increasing the full-load efficiency to 92.4%. Although the absolute efficiency improvement is only 0.3 percentage points, this is achieved under the premise of a significant improvement in reliability (junction temperature reduced by 15 degrees Celsius and ringing reduced by 52%), demonstrating the engineering value of joint optimization of "reliability and efficiency".

[0049] Comparative example (8.3ms sine half-wave) surge test ( (This refers to the rated forward single surge current of the TMBS device). The forward voltage drop (of the trench-type Schottky chip under rated forward current) rapidly rises to above 2.5V, causing localized overheating and chip burnout. In this invention, when a surge occurs, step S6 detects... It automatically raises the gate voltage to +3.0V, forming a strong inverse channel on the trench sidewall, effectively creating a parallel conduction path, thus... The voltage is reduced by approximately 0.8V, thereby increasing the surge withstand capability to 150A, a 25% improvement over the control, significantly enhancing the system's survivability under abnormal operating conditions such as power grid fluctuations or load changes.

[0050] The comparison results between the examples and the comparative examples are shown in the table below:

[0051] Therefore, the method of the present invention is significantly superior to the traditional fixed drive scheme in four aspects: suppressing voltage ringing, controlling junction temperature, improving efficiency, and enhancing surge capability. It achieves a win-win situation of "reliability improvement and efficiency optimization" and has outstanding substantive features and significant progress.

[0052] In summary, this adaptive rectification drive control method and system for trench-type Schottky chips utilizes the inherent characteristics of TMBS. As a temperature probe, the real-time junction temperature of the chip can be inferred by using a temperature drift model. This eliminates the need for expensive on-chip temperature sensors, transforming the device's "defect attributes" (leakage and temperature drift characteristics) into precise operational condition sensing resources; and through the corrected... Algorithm, introducing gate oxide thickness Dimensional correction was completed, and the critical electric field and parasitic inductance of the coupling medium were integrated. It also features leakage current and temperature drift characteristics. Under high temperature or high current conditions, it automatically limits the turn-off slope, effectively suppresses electric field concentration and voltage ringing at trench corners, significantly delays gate oxide aging, and extends device lifespan.

[0053] Through multidimensional coupling An adaptive algorithm automatically raises the gate voltage under heavy load and surge conditions, and utilizes the inversion layer on the trench sidewall to assist conduction, significantly reducing the device's forward voltage drop. Applying negative pressure grid offset under high temperature and light load conditions deepens the depletion layer on the trench sidewalls, effectively suppressing... The surge eliminates the risk of thermal runaway from a mechanistic perspective; it also constructs a two-way closed-loop linkage mechanism of "turn-off stress-conduction loss". When the turn-off speed is reduced and the switching loss is increased in order to suppress EMI ringing, the system automatically fine-tunes the gate voltage to compensate for the loss, breaking through the limitations of traditional single parameter optimization and achieving multi-objective global optimization of system efficiency, device reliability and EMI performance.

[0054] Moreover, this method is compatible with all TMBS and GCSD devices with independently led-out gates. It can be implemented simply by iterating the driver chip software algorithm without changing the chip process and packaging structure. It can be quickly deployed to adapt to various high-frequency rectifier power supply scenarios, with low upgrade costs and strong scalability.

[0055] The embodiments of the present invention are given for illustrative and descriptive purposes only, and are not intended to be exhaustive or to limit the invention to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of the invention, and to enable those skilled in the art to understand the invention and to design various embodiments with various modifications suitable for a particular purpose.

Claims

1. An adaptive rectification drive control method for trench-type Schottky chips, characterized in that: The trench-type Schottky chip is a trench-type MOS barrier Schottky diode with an independently led-out gate. Its trench gate is electrically connected to the external drive pin through a passivation layer opening. Includes the following steps: Step S1, Parameter pre-calibration: During the chip power-on initialization phase, the leakage current temperature sensitivity coefficient is obtained through temperature control testing. and voltage sensitivity And it is stored in non-volatile memory; Step S2: Within the turn-off steady-state window of each switching cycle, sample the anode voltage of the trench Schottky chip. With anode current From the anode current The reverse leakage current at the current moment is obtained by separation. ; Step S3, based on the reverse leakage current With current reverse bias Substituting the pre-stored leakage current temperature drift model, the junction temperature of the chip can be calculated. ; Step S4, the rate of change of voltage during the turn-off transient. In this stage, the equivalent junction capacitance under the current bias voltage is extracted online based on the displacement current formula. ; Step S5, set the junction temperature Reverse leakage current Equivalent junction capacitance Reverse bias Pre-stored parasitic inductance of power circuit and gate oxide thickness Substitute into the adaptive turn-off slope formula to calculate the target turn-off slope. and will The output is sent to the drive edge control module to regulate the shutdown process. ; Step S6, read the current load current Ringing amplitude in this cycle Combined with the results obtained in step S3 Substituting the result of step S5 into the gate bias adaptive formula, the trench gate bias voltage is calculated. ; Step S7, the... The output is sent to the trench gate drive stage, and during the forward conduction period, it is based on... Apply grid pressure to cause the trench sidewalls to form inverted auxiliary channels or deepen the depletion layer clamping as needed; Step S8, set the junction temperature for this cycle. Ringing amplitude and reverse leakage current Write the parameters into the linkage register as input parameters for steps S5 and S6 in the next cycle, forming a two-way closed-loop linkage control of the turn-off slope and gate bias.

2. The adaptive rectification drive control method for trench-type Schottky chips according to claim 1, characterized in that: The method for defining the turn-off steady-state window in step S2 is as follows: monitor the differential of the anode voltage after the turn-off edge ends. ,when hour, The maximum slope during the initial turn-off phase is defined as a time window lasting 50 ns to 200 ns; within this window, the anode current... The displacement current component in the middle approaches zero, therefore The reverse bias voltage Take the contents of this window The arithmetic mean.

3. The adaptive rectification drive control method for trench-type Schottky chips according to claim 1, characterized in that: The leakage current temperature drift model mentioned in step S3 is as follows: in, Reference leakage current; For reference junction temperature, the value is taken as follows. , The reference value under the rated reverse test voltage is determined by the temperature control test in step S1 and is used in step S1. Fit calibration; It is the base of the natural logarithm; The leakage current temperature sensitivity coefficient has a range of values. ; The leakage current voltage sensitivity is derived from the temperature control test fitting in step S1; The junction temperature, which is the unknown, can be obtained by transforming and inversely solving the above formula: 。 4. The adaptive rectification drive control method for trench-type Schottky chips according to claim 1, characterized in that: The adaptive turn-off slope formula mentioned in step S5 is: in The absolute turn-off slope is expressed in V / ns. , which is the withstand voltage of the dielectric layer at the trench corner, in volts (V). The critical breakdown electric field of the dielectric layer at the trench corner is derived from the derating value in the TCAD device simulation results. The physical thickness of the trench gate oxide dielectric layer is derived from chip manufacturing process design rules and is used to convert electric field strength into voltage dimensions. The reference voltage is the reverse bias voltage corresponding to the device's rated operating voltage. The value is in V and is used for dimension normalization; the rated operating voltage of the device is the maximum reverse operating voltage specified in the chip datasheet. The total parasitic inductance of the power circuit is derived from the parasitic parameter pre-characterization database of PCB layout and package structure; The transient peak current before the current cycle is turned off is derived from the anode Shunt resistor sampling circuit; The temperature drift-slope mapping coefficient has a value range of 0.5 to 2 and is calibrated by system loop stability simulation. 25℃ The reference leakage current under the given conditions is derived from the chip datasheet, where For measurement The reverse bias condition at that time is determined by taking the test voltage corresponding to the rated reverse working voltage at 25°C from the chip datasheet. This is the native shutdown slope of the driver chip under rated operating conditions, derived from the driver IC datasheet.

5. The adaptive rectification drive control method for trench-type Schottky chips according to claim 1, characterized in that: The online extraction described in step S4 The specific numerical calculation method is as follows: The integration interval is selected as 10% to 90% of the envelope of the turn-off transient voltage, i.e. , In the formula Used to eliminate the DC component of leakage current and retain the AC displacement current used for charging and discharging junction capacitance.

6. The adaptive rectification drive control method for trench-type Schottky chips according to claim 1, characterized in that: The gate offset adaptive formula mentioned in step S6 is: in, It serves as the zero bias or slight negative voltage reference for the trench gate, and its value is derived from the rated setting of the TMBS device structure. This is the current-gate gain coefficient, in V / A, calibrated by TCAD simulation, reflecting the increase in gate voltage and decrease in forward voltage drop. Marginal revenue; The surge-assisted trigger current threshold is set to 80% of the chip's rated current, derived from the device... The derating settings in the specification sheet; for The smoothing scaling factor of the function is used to prevent logic jitter. It comes from the system's anti-jitter design and its value ranges from 10% to 20% of the rated current. The leakage current penalty factor is the maximum allowable leakage current of the system. By reverse deduction, Derived from the system thermal design specifications; The ringing-loss compensation coefficient is determined by the overall system efficiency trade-off and has a value range of 1~2 V. The ringing amplitude detected in this cycle is derived from the peak sampling of the waveform turned off in step S5; The upper limit of the ringing amplitude allowed by the system is derived from EMI and device withstand voltage design specifications.

7. The adaptive rectification drive control method for trench-type Schottky chips according to claim 6, characterized in that: Step S6 also includes security protection logic: when a detection is made... At that time, the system will be forcibly locked. Any form of gate lifting operation is prohibited to prevent thermal runaway caused by leakage current at high temperatures.

8. The adaptive rectification drive control method for trench-type Schottky chips according to claim 1, characterized in that: The bidirectional closed-loop linkage mechanism described in step S8 specifically includes: Forward link: Output of step S3 and the output of step S5 Stored in the linkage register for calculation in step S6. The basis for the judgment at that time; Reverse link: Output of step S7 The width of the depletion region on the trench sidewall was altered, resulting in the extraction in the next cycle step S4. The value is offset, and this offset is used to monitor the aging status of the device or for advanced prediction algorithms.

9. The adaptive rectification drive control method for trench-type Schottky chips according to claim 1, characterized in that: The trench gate biasing in step S7 The scope of implementation is clamped to or Between; where positive voltage is used to form an inverse channel to reduce the forward voltage drop, and negative voltage is used to deepen depletion to suppress high-temperature leakage current; and 10ns before the turn-off signal is issued, the gate voltage is forcibly reset to or This eliminates turn-off noise introduced by gate-anode capacitive coupling; at the same time, during the conduction period, it ensures that the difference between the anode potential and the gate potential does not exceed the gate oxide tolerance limit.

10. An adaptive rectification drive control system adapted to trench-type Schottky chips, characterized in that: The system employs an adaptive rectification drive control method for trench-type Schottky chips as described in any one of claims 1-9, comprising: The parameter pre-calibration module is used to obtain the leakage current temperature sensitivity coefficient through temperature control testing during the chip power-on initialization phase. and voltage sensitivity And it is stored in non-volatile memory; The leakage current sampling module is used to sample the anode voltage of the trench Schottky chip within the turn-off steady-state window of each switching cycle. With anode current From the anode current The reverse leakage current at the current moment is obtained by separation. ; Junction temperature reverse calculation module, used to calculate based on the reverse leakage current With current reverse bias Substituting the pre-stored leakage current temperature drift model, the junction temperature of the chip can be calculated. ; Junction capacitance extraction module for voltage change rate during turn-off transients. In this stage, the equivalent junction capacitance under the current bias voltage is extracted online based on the displacement current formula. ; The turn-off slope calculation module is used to calculate the junction temperature. Reverse leakage current Equivalent junction capacitance Reverse bias Pre-stored parasitic inductance of power circuit and gate oxide thickness Substitute into the adaptive turn-off slope formula to calculate the target turn-off slope. and will The output is sent to the drive edge control module to regulate the shutdown process. ; The grid bias calculation module is used to read the current load current. Ringing amplitude in this cycle The junction temperature back-calculation module provides the following results: The results from the turn-off slope calculation module are substituted into the gate bias adaptive formula to calculate the trench gate bias voltage. ; Gate drive module, used to drive the gate drive module. The output is sent to the trench gate drive stage, and during the forward conduction period, it is based on... Apply grid pressure to cause the trench sidewalls to form inverted auxiliary channels or deepen the depletion layer clamping as needed; The closed-loop linkage register is used to set the junction temperature for the current cycle. Ringing amplitude and reverse leakage current The parameters are written into the linkage register and used as input parameters for the turn-off slope calculation module and the gate bias calculation module in the next cycle, forming a two-way closed-loop linkage control of the turn-off slope and the gate bias.

Images