A battery charger low temperature start method and system

Abstract

The application discloses a battery charger low-temperature starting method and system, and relates to the technical field of battery charging. The method comprises the following steps: acquiring the current environment temperature of a battery charger; according to the current environment temperature, reading a compensation parameter set matched with the current environment temperature from a pre-stored multi-dimensional lookup table, wherein the compensation parameter set at least comprises a target gate drive voltage value and a minimum conduction pulse width value; and driving a gallium nitride power switch tube to perform soft starting by using the target gate drive voltage value and the minimum conduction pulse width value. Through the temperature self-adaptive lookup table compensation mechanism, the application realizes accurate perception and effective inhibition of the deterioration of the low-temperature dynamic performance of the gallium nitride power device, significantly improves the starting success rate of the charger in a low-temperature environment, and guarantees the gate reliability and dynamic efficiency.

Description

Technical Field

[0001] This invention relates to the field of battery charger technology, and specifically to a method and system for low-temperature start-up of a battery charger. Background Technology

[0002] With the rapid popularization of consumer electronics, gallium nitride (GaN) power devices, with their advantages of high switching speed, low gate charge, and low on-resistance, are gradually replacing traditional silicon-based power devices and becoming the core power switching element of fast chargers. However, the operating characteristics of GaN power devices at low temperatures differ significantly from those at room temperature, posing a serious challenge to the low-temperature startup of chargers.

[0003] Existing technologies have addressed the temperature characteristics of GaN power devices. For example, some solutions provide lower gate-source voltages to drive GaN power transistors at low temperatures to avoid increased gate insulation failure due to excessively high gate-source voltages; others provide higher gate-source voltages to drive GaN power transistors at high temperatures to reduce on-resistance. These solutions, through gate voltage adjustment in a single temperature dimension, improve the gate reliability of GaN power transistors at low temperatures to some extent.

[0004] However, the aforementioned existing technologies and similar solutions still have the following shortcomings: Firstly, in low-temperature environments, GaN power devices not only face gate reliability issues, but also a more severe challenge: a significant exacerbation of the current collapse effect. At low temperatures, the trap charge capture rate inside the device increases, and the release time constant lengthens, resulting in a dynamic on-resistance Rds(on) much higher than the static nominal value. This leads to a sharp increase in conduction losses, insufficient output current during startup, and in severe cases, startup failure. Existing solutions only focus on the static adjustment of the gate-source voltage, without addressing the suppression of the current collapse effect.

[0005] Secondly, existing solutions rely on empirical compensation based on a single temperature dimension, neglecting several key physical dimensions affecting the dynamic characteristics of GaN devices—including drain static bias voltage and switching frequency. Studies have shown that the degree of mitigation of current collapse at low temperatures is closely related to the bias type, and the compensation requirements differ significantly under different bias conditions. Furthermore, while switching speed increases at low temperatures, the gate degradation rate actually accelerates. A single-dimensional compensation strategy cannot accurately match the actual operating state of the device, resulting in a 5%–10% deviation in the compensation parameters.

[0006] Third, in extremely low-temperature environments (such as below -40°C), the equivalent series resistance (ESR) of the electrolytic capacitors inside the charger increases sharply due to the low temperature. This may cause the controller's supply voltage to drop below the undervoltage lockout (UVLO) threshold at startup, forcing the controller to forcibly stop driving the output. At this time, even if the software algorithm is designed with a sophisticated compensation strategy, it cannot be executed because the controller cannot start normally. Existing technology does not provide an effective solution to this hardware limitation.

[0007] In summary, there is an urgent need in this field for a battery charger low-temperature startup method and system that can accurately sense the low-temperature dynamic characteristics of GaN devices, effectively suppress current collapse effects, and overcome hardware limitations in low-temperature startup. Summary of the Invention

[0008] The purpose of this invention is to provide a method and system for low-temperature start-up of a battery charger to solve the problems mentioned in the background art.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a low-temperature start-up method for a battery charger, wherein the battery charger includes a gallium nitride power switch and a digital controller, characterized in that the method includes: Obtain the current ambient temperature of the battery charger; Based on the current ambient temperature, a set of compensation parameters matching the current ambient temperature is read from a pre-stored multidimensional lookup table. The set of compensation parameters includes at least the target gate drive voltage value and the minimum turn-on pulse width value. The gallium nitride power switch is driven to perform a soft start using the target gate drive voltage value and the minimum conduction pulse width value.

[0010] The low-temperature start-up method for a battery charger according to the present invention includes a multi-dimensional lookup table, which is a three-dimensional lookup table whose index dimensions include ambient temperature, drain static bias voltage and switching frequency; the compensation parameter set also includes the expected dynamic on-resistance increment value.

[0011] The low-temperature start-up method for a battery charger according to the present invention, wherein the three-dimensional lookup table is pre-constructed through the following steps: The gallium nitride power switch was subjected to pulse IV testing under full temperature range and full bias conditions to obtain the static on-resistance at each temperature point. The dynamic on-resistance increment at each temperature point and under each bias condition and the threshold voltage at each temperature point ; The test data is fitted into a compressed empirical formula and then stored in non-volatile memory in the form of a three-dimensional lookup table.

[0012] The low-temperature start-up method for a battery charger according to the present invention, wherein the target gate drive voltage value The following equation must be satisfied: ; in, This is the nominal gate drive voltage at room temperature. This is the gate voltage correction amount that increases as the temperature decreases; and Not lower than the preset lower limit value The lower limit value The minimum gate voltage required to ensure reliable conduction of the gallium nitride power switch.

[0013] The low-temperature start-up method for a battery charger according to the present invention, wherein the minimum conduction pulse width value The following equation must be satisfied: ; in The effective release time constant of the trap charge is obtained through offline calibration and is related to the current ambient temperature and the drain static bias voltage; k is a design coefficient with a value range of 3≤k≤5.

[0014] The low-temperature start-up method for a battery charger according to the present invention includes a target gate drive voltage value and a minimum conduction pulse width value stored in a multidimensional lookup table. When the current ambient temperature is lower than a first preset threshold, the multidimensional lookup table outputs a reduced target gate drive voltage value and an extended minimum conduction pulse width value. The reduced target gate drive voltage value and the extended minimum conduction pulse width value are read from the same lookup table entry based on the same current ambient temperature and drain static bias voltage.

[0015] The low-temperature start-up method for a battery charger according to the present invention further includes a pre-start-up self-test step before performing the soft start: The digital controller sends a test pulse to drive the gallium nitride power switch and monitors the voltage drop of the digital controller's power supply. Based on the voltage drop, a target start position is selected from multiple preset start positions; the multiple preset start positions include at least a full-power fast start position and a reduced-power slow start position.

[0016] The low-temperature start-up method for a battery charger according to the present invention includes a preheating mode when the power supply voltage drops by more than a second preset threshold. The digital controller then enters a preheating mode by emitting a narrow pulse at a drive frequency lower than the normal operating frequency, generating heat by utilizing the switching losses of the gallium nitride power switch. Once the current ambient temperature rises to above a third preset threshold, the preheating mode is exited and the soft start is executed.

[0017] The low-temperature start-up method for a battery charger according to the present invention includes a soft start-up comprising two-stage start-up steps: First-level startup: The digital controller performs only temperature readings and basic initialization, operating in an ultra-low power mode; Once the power supply to the digital controller is confirmed to be stable, the second-level startup is executed: the multidimensional lookup table is loaded and the main control algorithm is started, outputting the drive signal corresponding to the compensation parameter set.

[0018] Furthermore, the present invention also provides a battery charger low-temperature start-up system for performing the above-described battery charger low-temperature start-up method. This system includes: Gallium nitride power switches are used as the main power switching components in battery chargers. Temperature sensing module, used to collect the current ambient temperature in real time; Non-volatile memory is used to store multidimensional lookup tables. The index dimensions of the multidimensional lookup tables include ambient temperature, drain static bias voltage and switching frequency. Each entry stores the target gate drive voltage value, the minimum on-pulse width value and the expected dynamic on-resistance increment value. A digital controller is connected to the temperature sensing module, the non-volatile memory, and the gallium nitride power switch, respectively. A power supply circuit is used to provide operating voltage to the digital controller, and the power supply circuit includes low-temperature redundant capacitor elements.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows: By employing the core steps of "obtaining the current ambient temperature—reading the compensation parameter set from a multi-dimensional lookup table—driving the gallium nitride power switch to perform soft start using the compensation parameters," a temperature-adaptive lookup table compensation mechanism is introduced for the first time into the low-temperature start-up control of gallium nitride chargers. This achieves automatic adaptation of the drive strategy to the low-temperature environment, eliminating the failure risk of fixed-parameter drives at low temperatures. Furthermore, the compensation parameter set simultaneously includes the target gate drive voltage value and the minimum conduction pulse width value, suppressing the degradation of low-temperature dynamic performance from both the voltage and time domains. This provides superior compensation compared to existing technologies that only focus on adjusting the single parameter of the gate voltage. Simultaneously, the soft-start execution method avoids damage to the gallium nitride power switch caused by the instantaneous current surge during startup, improving the success rate of low-temperature startup while ensuring device safety. Attached Figure Description

[0020] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the method steps of the present invention.

[0022] Figure 2 This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0023] The terms "first," "second," "third," and "fourth," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0024] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0025] "Multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0026] Furthermore, the terms indicating orientation, such as "up," "down," "left," "right," "upper end," "lower end," and "longitudinal," are all based on the posture and position of the device or equipment described in this solution during normal use.

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, a clear and complete description will be provided below in conjunction with the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0028] like Figure 1-2 As shown, the present invention provides a low-temperature start-up system for a battery charger, comprising a gallium nitride (GaN) power switch, a temperature sensing module, a non-volatile memory, a digital controller, and a power supply circuit. The GaN power switch, serving as the main power switching element of the battery charger, is a 650V enhancement-mode GaNHEMT device with a threshold voltage Vgs(th) of approximately 1.7V to 2.1V at room temperature, a gate withstand voltage of +6.5V, and an on-resistance Rds(on) of 160mΩ to 225mΩ.

[0029] The temperature sensing module includes a temperature sensor, which is an NTC thermistor with an operating temperature range of -55°C to +125°C. At 25°C, its resistance is 10kΩ with a resistance tolerance of ±1% and a B value of 3987K. It can acquire the current ambient temperature in real time with an accuracy better than ±1°C.

[0030] The non-volatile memory uses EEPROM or flash memory chips to store the multidimensional lookup table, with a capacity of not less than 32KB, to ensure the complete storage of the three-dimensional lookup table and firmware program.

[0031] The digital controller uses a digital power control chip with ultra-low startup current characteristics, with a typical startup current of no more than 10µA and an operating temperature range of -40°C to +125°C. It integrates an analog-to-digital converter (ADC) for reading analog signals from the temperature sensor, a pulse width modulation (PWM) generator for driving gallium nitride power switches, and a read / write interface for non-volatile memory.

[0032] The digital controller is connected to the temperature sensor, non-volatile memory, and gallium nitride power switch. The power supply circuit provides the operating voltage VCC to the digital controller and includes a low-temperature redundant capacitor element. This low-temperature redundant capacitor element is a conductive polymer solid-state capacitor with a capacitance of 47µF and a rated voltage of 16V. Its equivalent series resistance (ESR) remains stable below 20mΩ over a wide temperature range of -55°C to +105°C. Compared to the drawback of traditional aluminum electrolytic capacitors where the ESR increases sharply at low temperatures, the conductive polymer solid-state capacitor maintains almost a constant ESR at low temperatures, effectively preventing the supply voltage VCC from dropping below the undervoltage lockout threshold at low temperatures.

[0033] like Figure 1 As shown, the low-temperature start-up control process of the present invention includes main steps such as temperature acquisition, parameter reading, soft start execution, and dynamic closed-loop adjustment.

[0034] In practice, the low-temperature start-up method for battery chargers includes the following steps.

[0035] First, obtain the current ambient temperature T of the battery charger.

[0036] The digital controller reads the voltage across the temperature sensor (NTC thermistor) using its internal analog-to-digital converter and calculates the current ambient temperature T based on the NTC thermistor's resistance-temperature characteristic curve. When the ambient temperature T is within the normal temperature range (T≥-10°C), the digital controller drives the gallium nitride power switch according to the nominal parameters at normal temperature; when the ambient temperature T is below -10°C, the digital controller determines that it is entering the low-temperature start-up mode.

[0037] Secondly, based on the current ambient temperature T, the digital controller reads a set of compensation parameters matching the current ambient temperature T from a multidimensional lookup table pre-stored in non-volatile memory. This set of compensation parameters includes at least the target gate drive voltage value. and minimum conduction pulse width value This multidimensional lookup table is a three-dimensional lookup table, and its index dimensions include ambient temperature T and drain static bias voltage. and switching frequency Each entry stores the target gate drive voltage value. Minimum conduction pulse width value and the expected dynamic on-resistance increment value The digital controller locates the corresponding entry in the three-dimensional lookup table and reads all compensation parameters based on the current real-time monitored drain voltage and the set switching frequency.

[0038] Finally, the digital controller uses the drive signal output from its internal PWM generator, with the target gate drive voltage value Vgs,opt and the minimum on-pulse width value... The GaN power switch is driven to perform a soft start. During the soft start process, the digital controller starts from zero duty cycle and gradually increases the duty cycle of the PWM signal according to a preset step rate. After each step, at each stage of the soft start, the digital controller collects the on-state voltage drop V0 of the GaN power switch in real time through current sampling circuits and voltage sampling circuits. ds_on and conduction current I ds Calculate the actual dynamic on-resistance R ds_dyn_actual =V ds_on / I ds Then, this actual value is compared with the expected dynamic on-resistance increment value ΔR stored in the three-dimensional lookup table.ds_exp Compare them.

[0039] If the actual value is within ±10% of the expected value, the current compensation parameter is determined to be accurate and effective, and subsequent soft-start steps are continued. If the actual value deviates from the expected value by more than +10% but less than +30%, the device's current dynamic characteristics are determined to have deteriorated compared to the calibrated state (possibly due to device aging or batch differences). In the next soft-start step, the digital controller automatically adjusts the minimum conduction pulse width. Extend the length by 20% to compensate for the increased dynamic on-resistance; if the actual value deviates from the expected value by more than -10% but less than -30% (actual value is less than the expected value by 10% to 30%), then the device's current dynamic characteristics are considered better than the calibration state, and the length can be appropriately shortened. To improve efficiency; if the actual value deviates from the expected value by more than ±30% (the deviation exceeds 30%), it is determined that there is an abnormality in the device or system, and the startup is immediately terminated and a fault alarm is triggered. This mechanism upgrades "open-loop lookup table compensation" to "closed-loop lookup table compensation plus real-time verification", so that the compensation strategy can be dynamically adapted according to the actual health status of the device, rather than relying entirely on the static expectations calibrated at the factory, and then proceeding to the next step of increasing the duty cycle. Through the above real-time health status assessment, this invention combines the static expectations calibrated at the factory with the actual feedback during the startup process, realizing the dynamic adaptation of compensation parameters to individual differences and aging states of devices.

[0040] The above multidimensional lookup table is constructed as follows.

[0041] First, pulse IV tests were performed on the gallium nitride power switch under full temperature range and full bias conditions.

[0042] Specifically, the gallium nitride power switch was placed in a temperature chamber with a temperature scan range of -55°C to +25°C and a step size of 5°C; at each temperature point, a different drain static bias voltage was set. (From 0V to 650V, distributed in logarithmic intervals) and different switching frequencies (Distributed at octave intervals from 50kHz to 500kHz); The static on-resistance under various conditions was measured using the pulse IV test method. Dynamic on-resistance and threshold voltage .

[0043] Then, the test data is fitted to a compressed empirical formula: (Equation 1); in, The static on-resistance at temperature T is the on-resistance of the gallium nitride power switch at a given ambient temperature T without any switching action. This value decreases slightly as the temperature decreases. β(T) is a temperature-dependent current collapse intensity coefficient, dimensionless, reflecting the enhancement of dynamic on-resistance by the trapped charge trapping effect at a given temperature T. This coefficient increases as temperature decreases; β(T) is a temperature-dependent bias voltage exponent coefficient, dimensionless, reflecting the drain static bias voltage. The degree of nonlinear influence on the current collapse effect is obtained by fitting experimental data; This is a temperature-dependent frequency attenuation coefficient, measured in Hz, reflecting the switching frequency. Regarding the saturation characteristics of dynamic on-resistance, when much smaller The dynamic resistance increment increases linearly with frequency, when Much larger The dynamic resistance increment tends to saturate, and this coefficient decreases as the temperature decreases, indicating that the response speed of the trapped charge is slower and the frequency saturation point is lower at low temperatures. This is the drain static bias voltage, which is the voltage between the drain and source of the gallium nitride power switch in the off state, in volts (V). is the switching frequency, i.e., the number of times the gallium nitride power switch turns on and off per second, in Hz; e is the base of the natural logarithm (Euler's number), approximately 2.71828. The essence of this compressed empirical formula is to describe discrete test data continuously through a function driven by physical mechanisms, where the exponential term... The power-law dependence of the drain bias voltage on the trap charge capture probability is described, while The dynamic saturation effect of switching frequency on trap charge accumulation is described.

[0044] The fitted model parameters (i.e., at each temperature point) , , , The data is stored in non-volatile memory in the form of a three-dimensional lookup table. During table lookup, the digital controller determines the data based on the current temperature T and the drain static bias voltage. and switching frequency In the 3D lookup table, if the current parameter falls exactly on the calibration point, it is read directly; otherwise, if it falls between calibration points, linear interpolation or spline interpolation is used to calculate the accurate value.

[0045] During low-temperature startup, the target gate drive voltage value The determination method is as follows.

[0046] The digital controller reads the current temperature T and drain static bias voltage from a three-dimensional lookup table in non-volatile memory. The corresponding target gate drive voltage value satisfies the following relationship: (Equation 2); in, The nominal gate drive voltage at room temperature is 6.5V in this embodiment. This is the recommended gate drive voltage value for gallium nitride power switches to achieve the lowest on-resistance at room temperature. This is the gate voltage correction amount that increases as the temperature decreases, expressed in V. Its physical meaning is the voltage value that needs to be subtracted from the nominal gate voltage to suppress the accelerated degradation of the gate insulating layer at low temperatures.

[0047] Specifically, The rule for determining the value is: when T ≥ -10°C =0V; when -30°C ≤ T < -10°C =0.5V; when -40°C≤T<-30°C =1.0V; when T<-40°C =1.5V. Meanwhile, based on the drain static bias voltage... Make fine adjustments: when For voltages ≥300V, reduce the voltage drop by 0.2V from the aforementioned baseline value to maintain low on-resistance under high bias conditions. Furthermore... Not lower than the preset lower limit value The lower limit value To ensure the minimum gate voltage required for reliable conduction of the gallium nitride power switch, 4.5V is used in this embodiment. By actively reducing the gate drive voltage at low temperatures, excessively high gate voltages are avoided from accelerating the failure of the gate insulating layer (experiments show that if the nominal gate voltage is maintained at -55°C, the gate insulating layer lifetime may be shortened to about 20 hours, while reducing the gate voltage to the range of 4.5V to 5.5V can extend the gate lifetime to more than 1000 hours).

[0048] At the same time, the minimum conduction pulse width value The determination method is as follows.

[0049] The digital controller reads the minimum on-pulse width value corresponding to the current temperature T and the drain static bias voltage VdsQ from a three-dimensional lookup table in non-volatile memory. This value satisfies the following relationship: (Equation 3); in, The effective release time constant of the trapped charge, in seconds (s), is obtained through offline calibration and is related to the current ambient temperature T and the drain static bias voltage. Related; its physical meaning is the characteristic time required for a trapped charge to be released from the trap under given temperature and bias conditions. At low temperatures, the trapping process of the trapped charge is suppressed, while the time constant of the emission process increases significantly; at a temperature of 150K (approximately -123°C), the release time constant of the trapped charge can reach tens of seconds or more. In this embodiment, Typical values ​​range from 0.2µs to 0.5µs at room temperature, from 0.6µs to 1.5µs at -30°C, and from 1.0µs to 3.0µs at -40°C, under high drain bias conditions. ≥300V) The value is further increased by 20% to 50%. k is a design coefficient, ranging from 3 ≤ k ≤ 5. Its physical meaning is to ensure that the conduction pulse width reaches 3 to 5 times the trap charge release time constant, thus allowing sufficient time for the trap charge to release during each conduction period, avoiding continuous deterioration of the dynamic on-resistance due to cycle-by-cycle accumulation. The digital controller determines the dynamic on-resistance based on the current temperature T and the drain static bias voltage. Read from the 3D lookup table Multiplying by the design factor k yields the minimum conduction pulse width value. If the current load demand requires a certain turn-on time... Less than The digital controller will then force the actual conduction time to be set to... The switching frequency or duty cycle is adjusted accordingly to maintain constant output power.

[0050] Target gate drive voltage value With minimum conduction pulse width value Linked storage is used in a three-dimensional lookup table. Specifically, in a three-dimensional lookup table of non-volatile memory, each entry simultaneously stores the data under that operating condition. and When the current ambient temperature T is lower than the first preset threshold (in this embodiment, the first preset threshold is -10°C), the target gate drive voltage value read by the digital controller from the three-dimensional lookup table is a reduced value (lower than the nominal value of 6.5V at room temperature), and the minimum conduction pulse width value read is an extended value (extended than the nominal value at room temperature). The reduced target gate drive voltage value and the extended minimum conduction pulse width value are based on the same current ambient temperature T and drain static bias voltage. Read from the same lookup table entry.

[0051] This linked storage mechanism ensures that the gate voltage compensation strategy and the pulse width optimization strategy are adjusted in tandem under identical operating conditions, avoiding potential strategy conflicts that might result from separate adjustments—for example, if the gate voltage compensation algorithm is based on temperature T1 and bias... The voltage drop value is calculated, and the pulse width optimization algorithm is based on temperature T2 and bias. The calculated delay value may be mismatched due to different operating conditions. Linked storage fundamentally eliminates this risk.

[0052] Before performing a soft start, the digital controller also performs a pre-start self-test.

[0053] Specifically, the digital controller first sends a test pulse with a width of less than 100ns to drive the gallium nitride power switch. At the same time, the digital controller monitors the drop in voltage VCC from the power supply circuit to the digital controller in real time through its internal voltage monitoring circuit. The width of this test pulse is chosen to be less than 100ns in order to generate a measurable drop in the power supply voltage while avoiding damage to the gallium nitride power switch from entering a high-current conduction state.

[0054] The digital controller selects the target start position from multiple preset start positions based on the voltage drop ΔVCC. The multiple preset start positions include at least a full-power fast start position and a reduced-power soft start position, wherein the duty cycle step rate of the full-power fast start position is 5% per millisecond, and the duty cycle step rate of the reduced-power soft start position is 1% per millisecond.

[0055] The specific selection rules are as follows: If ΔVCC < 5%VCC nom (VCC) nom If the nominal output voltage of the power supply circuit is 12V (in this embodiment, it is taken as 12V), then the full power fast start mode is selected; If 5% VCC nom ≤ΔVCC<15%VCC nom Then select the reduced power soft start mode; If ΔVCC ≥ 15%VCC nom If so, it is determined that the power supply capacity is severely insufficient.

[0056] The second preset threshold is 15% VCC. nom When the supply voltage drop ΔVCC exceeds the second preset threshold (15%VCC)... nom When the digital controller enters the preheating mode, it sends a narrow pulse with a width of ton,min to drive the gallium nitride power switch at a drive frequency lower than the normal operating frequency (1 / 10 of the normal operating frequency in this embodiment, i.e., 5kHz to 10kHz). The heat generated by the switching loss of the gallium nitride power switch is used to heat the internal environment of the charger.

[0057] In preheating mode, since the driving frequency is much lower than the normal operating frequency, the average switching loss is controlled below 10% of the normal operating loss, which will not cause additional burden on the power supply circuit. The digital controller continuously monitors the current ambient temperature T through a temperature sensor. After the current ambient temperature T rises back to above the third preset threshold (in this embodiment, the third preset threshold is -10°C), the digital controller exits the preheating mode and executes the normal soft start process.

[0058] Furthermore, upon initial low-temperature startup failure, the digital controller records the operating parameters at the time of failure, including the current ambient temperature T, the supply voltage drop magnitude ΔVCC, and the duty cycle D at the time of failure. The digital controller further selects different recovery paths based on the type of failure trigger source—if the failure is triggered by undervoltage lockout (i.e., the supply voltage VCC drops below the UVLO threshold), it determines that the power supply capability is severely insufficient and enters a preheating mode to increase the ambient temperature and reduce the electrolytic capacitor ESR; if the failure is triggered by overcurrent protection (i.e., the startup current exceeds the preset overcurrent threshold), it determines that the current gate drive voltage is too high or the duty cycle step rate is too fast, maintaining a voltage not lower than the minimum required for the current gate drive. Under the premise of further reducing the gate drive voltage and halving the duty cycle step rate, the system re-attempts startup. If the failure is triggered by timeout protection (i.e., the soft-start process exceeds the preset maximum allowable time), it is determined that the current compensation parameters are too conservative, resulting in a slow startup speed. Under the premise of ensuring safety, the duty cycle step rate is appropriately increased to shorten the startup time. Through this failure cause-oriented hierarchical recovery strategy, the system can take differentiated measures for different failure mechanisms, avoiding repeated failures caused by general retrying, and significantly improving the cumulative startup success rate after multiple attempts.

[0059] In addition, the soft boot process includes two boot steps.

[0060] First-stage startup: After the battery charger is powered on, the digital controller initially performs only temperature reading and basic initialization operations, including reading the current ambient temperature T from the temperature sensor via the analog-to-digital converter, initializing internal registers, and configuring the basic clock. During this stage, the digital controller operates in ultra-low power mode, with its operating current not exceeding 50µA. This ensures that even under extremely low temperature conditions where the power supply circuit's load capacity is severely limited, the digital controller can still reliably start up and complete temperature sensing. Once the power supply to the digital controller is confirmed to be stable (i.e., the digital controller detects that the supply voltage VCC is stable within ±5% of VCC_nom for 10 consecutive sampling cycles (each sampling cycle is 1ms), the second-stage startup is executed: The digital controller loads a three-dimensional lookup table from non-volatile memory into its internal cache, initiates the main control algorithm (including gate voltage compensation algorithm, pulse width optimization algorithm, pre-start self-test algorithm, etc.), and adjusts the current ambient temperature T and drain static bias voltage accordingly. and switching frequency The compensation parameter set is read from a three-dimensional lookup table, and the corresponding drive signal is output to the gallium nitride power switch. The two-stage startup mechanism ensures that during the initial startup phase when the power supply is weakest, the digital controller completes the most basic temperature sensing and status judgment with minimal power consumption, creating the prerequisite for loading the complete algorithm and executing the compensation drive.

[0061] Through the coordinated action of the above steps, this invention achieves precise sensing and effective suppression of the low-temperature dynamic performance degradation of gallium nitride power switches. The three-dimensional lookup table incorporates ambient temperature T and drain static bias voltage. and switching frequency With three index dimensions, the model prediction error is controlled within 5%, solving the problem that the model accuracy is insufficient in the existing technology and therefore cannot accurately perceive the device status.

[0062] The gate voltage compensation mechanism actively reduces the gate drive voltage at low temperatures. This avoids excessively high gate voltage accelerating gate insulation layer failure, extending gate lifetime at extremely low temperatures from approximately 20 hours to over 1000 hours. The pulse width optimization mechanism extends the minimum conduction pulse width. Sufficient release time is provided for the trapped charge, keeping the increase in dynamic on-resistance at low temperatures below 12% from 40%~60%. A dual-parameter linked storage mechanism ensures that gate voltage compensation and pulse width optimization are adjusted collaboratively under the same operating conditions, avoiding strategy conflicts.

[0063] Pre-start self-test, tiered start-up selection, preheating mode, and two-stage start-up mechanisms overcome the hardware limitations at the software level, such as the increased ESR of electrolytic capacitors at extremely low temperatures leading to a drop in supply voltage and subsequent undervoltage lockout (UVLO) failure. This improves the success rate of low-temperature start-up at -40°C to over 99%. Conductive polymer solid-state capacitors, as low-temperature redundant capacitors in the power supply circuit, maintain almost unchanged ESR across a wide temperature range of -55°C to +105°C, providing a stable power supply foundation for the software algorithm at the hardware level. These hardware and software redundancy safeguards together form a complete technical closed loop of "accurate sensing → effective compensation → reliable execution."

[0064] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for low-temperature start-up of a battery charger, the battery charger comprising a gallium nitride power switch and a digital controller, characterized in that, The method includes: Obtain the current ambient temperature of the battery charger; Based on the current ambient temperature, a set of compensation parameters matching the current ambient temperature is read from a pre-stored multidimensional lookup table. The set of compensation parameters includes at least the target gate drive voltage value and the minimum turn-on pulse width value. The gallium nitride power switch is driven to perform a soft start using the target gate drive voltage value and the minimum conduction pulse width value.

2. The low-temperature start-up method for a battery charger according to claim 1, characterized in that, The multidimensional lookup table is a three-dimensional lookup table, and its index dimensions include ambient temperature, drain static bias voltage, and switching frequency; the compensation parameter set also includes the expected dynamic on-resistance increment value.

3. The low-temperature start-up method for a battery charger according to claim 2, characterized in that, The three-dimensional lookup table is pre-constructed through the following steps: The gallium nitride power switch was subjected to pulse IV testing under full temperature range and full bias conditions to obtain the static on-resistance at each temperature point. The dynamic on-resistance increment at each temperature point and under each bias condition and the threshold voltage at each temperature point ; The test data is fitted into a compressed empirical formula and then stored in non-volatile memory in the form of a three-dimensional lookup table.

4. The low-temperature start-up method for a battery charger according to claim 1, characterized in that, The target gate drive voltage value The following equation must be satisfied: ; in, This is the nominal gate drive voltage at room temperature. This is the gate voltage correction amount that increases as the temperature decreases; and Not lower than the preset lower limit value The lower limit value The minimum gate voltage required to ensure reliable conduction of the gallium nitride power switch.

5. The low-temperature start-up method for a battery charger according to claim 2, characterized in that, The minimum conduction pulse width value The following equation must be satisfied: ； in The effective release time constant of the trap charge is obtained through offline calibration and is related to the current ambient temperature and the drain static bias voltage; k is a design coefficient with a value range of 3≤k≤5.

6. The low-temperature start-up method for a battery charger according to claim 4 or 5, characterized in that, The target gate drive voltage value and the minimum conduction pulse width value are stored together in the multidimensional lookup table. When the current ambient temperature is lower than a first preset threshold, the multidimensional lookup table outputs a reduced target gate drive voltage value and an extended minimum conduction pulse width value. The reduced target gate drive voltage value and the extended minimum conduction pulse width value are read from the same lookup table entry based on the same current ambient temperature and drain static bias voltage.

7. The low-temperature start-up method for a battery charger according to claim 1, characterized in that, Before performing the soft boot, a pre-boot self-test step is also included: The digital controller sends a test pulse to drive the gallium nitride power switch and monitors the voltage drop of the digital controller's power supply. Based on the voltage drop, a target start position is selected from multiple preset start positions; the multiple preset start positions include at least a full-power fast start position and a reduced-power slow start position.

8. The low-temperature start-up method for a battery charger according to claim 7, characterized in that, When the power supply voltage drops below the second preset threshold, the digital controller enters a preheating mode: it emits a narrow pulse at a drive frequency lower than the normal operating frequency, generates heat by utilizing the switching losses of the gallium nitride power switch, and exits the preheating mode and performs the soft start after the current ambient temperature rises above the third preset threshold.

9. The low-temperature start-up method for a battery charger according to claim 1, characterized in that, The soft boot process includes two boot steps: First-level startup: The digital controller performs only temperature readings and basic initialization, operating in an ultra-low power mode; Once the power supply to the digital controller is confirmed to be stable, the second-level startup is executed: the multidimensional lookup table is loaded and the main control algorithm is started, outputting the drive signal corresponding to the compensation parameter set.

10. A battery charger low-temperature start-up system for performing the battery charger low-temperature start-up method as described in any one of claims 1 to 9, characterized in that, include: Gallium nitride power switches are used as the main power switching components in battery chargers. Temperature sensing module, used to collect the current ambient temperature in real time; Non-volatile memory is used to store multidimensional lookup tables. The index dimensions of the multidimensional lookup tables include ambient temperature, drain static bias voltage and switching frequency. Each entry stores the target gate drive voltage value, the minimum on-pulse width value and the expected dynamic on-resistance increment value. A digital controller is connected to the temperature sensing module, the non-volatile memory, and the gallium nitride power switch, respectively. A power supply circuit is used to provide operating voltage to the digital controller, and the power supply circuit includes low-temperature redundant capacitor elements.

Images