A hall sensor adaptive initialization system and method based on closed-loop feedback

Abstract

This invention discloses an adaptive initialization system and method for a Hall sensor based on closed-loop feedback, belonging to the fields of integrated circuit design and magnetic sensor technology. The system is integrated within the Hall sensor integrated circuit and includes a Hall semiconductor sensing unit, a microcoil circuit unit, a microcoil driving circuit unit, a compensation and signal conditioning circuit unit, an analog-to-digital converter unit, a power management unit, and an initialization control logic unit. In initialization mode, the initialization control logic unit continuously acquires the Hall output value, compares it with a preset reference value to generate an error signal, and dynamically adjusts the direction and magnitude of the microcoil current through closed-loop feedback until the Hall output converges within the tolerance range of the reference value. This invention requires no external microprocessor intervention, realizing on-chip closed-loop adaptive initialization across the entire "sensing-decision-execution" chain. It supports multiple modes such as automatic power-on, external enable, and over-limit re-triggering, and has advantages such as high initialization accuracy, fast response speed, low system resource consumption, and ease of industrialization. It can be widely applied in automotive electronics, industrial servo systems, and low-cost IoT devices.

Description

Technical Field

[0001] This invention belongs to the field of integrated circuit design and magnetic sensor technology, specifically relating to a Hall sensor integrated circuit, and more particularly to a closed-loop feedback Hall sensor initialization system and method with adaptive initialization capability and no need for external microprocessor intervention. Background Technology

[0002] Hall effect sensors are widely used in motor control, position detection, current sensing, and industrial automation due to their advantages such as non-contact operation, high reliability, and low power consumption. In practical applications, Hall effect sensors often face complex and variable background magnetic field environments, especially when installed near strong magnetic equipment such as motors and transformers. External interference magnetic fields can directly affect the static operating point of the Hall element, causing the output signal to deviate from the ideal value, thereby affecting the system's measurement accuracy and response speed.

[0003] To eliminate the influence of background magnetic fields, existing technologies include programmable Hall sensor architectures with reverse magnetization capabilities. For example, a published Chinese invention patent application (application number: 2024107114755, publication number: CN118501782A, invention title: A Hall sensor integrated circuit for initialization using reverse magnetization) proposes an open-loop initialization scheme involving an external microprocessor (MCU). The specific process is as follows: the external MCU reads the Hall output voltage value via an I2C / SPI interface → the MCU internally calculates compensation parameters → the MCU then writes control commands to the sensor via a communication interface to adjust the micro-coil current to generate a reverse magnetic field. While this scheme can achieve a certain degree of environmental magnetic field cancellation, it still has shortcomings in automatically determining the reverse magnetic field parameters. Dependence on external processor: The initialization logic relies entirely on an external MCU, which increases the software burden and code complexity of the system's main controller; Open-loop control with limited convergence accuracy: Due to the lack of an output feedback mechanism, the external MCU cannot know the actual Hall output state after the reverse magnetic field is applied in real time, making it difficult to achieve accurate and dynamic closed-loop adjustment; Communication latency and resource consumption: I2C / SPI communication introduces additional latency, and in system startup or multi-sensor scenarios, communication bandwidth and MCU resources are significantly consumed, which is not conducive to applications with fast response and high real-time requirements; Difficulty in meeting high autonomy requirements: Automotive electronics, Industry 4.0 and low-cost IoT devices place higher demands on the "power-on and use" and "autonomous calibration" capabilities of sensors, which existing solutions cannot meet. Summary of the Invention

[0004] The present invention aims to solve the technical problems of existing Hall sensors with reverse magnetization function, which require an external microprocessor for parameter calculation and command issuance during the initialization process, and whose control process is open-loop, resulting in slow response and high system complexity.

[0005] Specifically, the core technical problems to be solved by this invention include: How to solve the problem of incorrect sensor readings when the environment of a Hall sensor becomes magnetized after prolonged use, resulting in a non-zero static magnetic field? How to enable the Hall sensor to autonomously determine the compensation current parameters required for the reverse magnetic field on-chip without the need for external MCU intervention; How to achieve closed-loop feedback control, dynamically monitor Hall output and adjust micro-coil excitation in real time until the output converges to the target value; How to achieve initialization upon power-on and on-demand initialization during operation; How to support multiple triggering methods (automatic power-on, external enable, threshold over-limit) to improve system integration and intelligence.

[0006] In view of this, the present invention aims to provide a Hall sensor adaptive initialization system and method that can autonomously complete reverse magnetic field initialization without external MCU intervention and has closed-loop dynamic adjustment capability, so as to overcome the above-mentioned shortcomings of the prior art.

[0007] To address the aforementioned technical problems, this invention discloses a Hall sensor adaptive initialization system and method based on closed-loop feedback, belonging to the fields of integrated circuit design and magnetic sensor technology. The system is integrated within the Hall sensor integrated circuit and includes a Hall semiconductor sensing unit, a micro-coil circuit unit, a micro-coil driving circuit unit, a compensation and signal conditioning circuit unit, an analog-to-digital converter unit, a power management unit, and an initialization control logic unit. In initialization mode, the initialization control logic unit continuously acquires the Hall output voltage, compares it with a preset reference value to generate an error signal, and dynamically adjusts the direction and magnitude of the micro-coil voltage and current through closed-loop feedback until the Hall output converges within the reference value tolerance range. This invention requires no external microprocessor intervention, achieving on-chip closed-loop adaptive initialization across the entire "sensing-decision-execution" chain. It supports multiple modes such as automatic power-on, external enable, and over-limit re-triggering, and has advantages such as high initialization accuracy, fast response speed, low system resource consumption, and ease of industrialization. It can be widely applied in automotive electronics, industrial servo systems, and low-cost IoT devices.

[0008] The system specifically includes the following components: Hall semiconductor sensing unit (1): used to sense the external magnetic field to be measured and the environmental interference magnetic field, and output a differential voltage signal; Micro-coil circuit unit (2): Integrated inside the chip, located near the Hall sensing unit, used to generate an adjustable direction and intensity reverse compensation magnetic field under current excitation; Micro-coil drive circuit unit (3): Receives control signals from the initialization control logic unit of the analog-to-digital converter unit and adjusts the direction and magnitude of the voltage and current flowing through the micro-coil; Compensation and signal conditioning circuit unit (4): amplifies, temperature compensates and filters the original signal output by the Hall sensing unit to improve the signal-to-noise ratio and stability; Analog-to-digital converter unit (5): converts the conditioned analog Hall output voltage into a digital quantity for on-chip logic processing. The analog-to-digital converter contains a window comparator. During the initialization process, when the initialization result is 0 or within the permissible error range, it outputs a GOOD signal. Power management unit (6): provides a stable, ultra-low noise power supply voltage for the above modules and generates a power-on reset (RESET) signal; Initialization control logic unit (7): It is connected to the analog-to-digital converter unit (5) and the micro-coil drive circuit unit (3) respectively, and is the core control module for realizing closed-loop adaptive initialization in this invention.

[0009] Specifically, the initialization control logic unit (7) also includes an initialization control port, which is implemented using an I2C interface for optional interaction with an external controller (such as an MCU). The I2C interface supports the following functions: receiving "initialization on demand" trigger instructions from an external MCU (e.g., by writing to specific register bits), and optionally providing an initialization status flag, such as "initialization complete". However, the I2C interface does not participate in parameter calculation or current adjustment instruction transmission in the closed-loop feedback control process, that is, the external MCU does not need to read the Hall value, calculate compensation parameters, or send adjustment instructions to the micro-coil drive circuit unit. All closed-loop control logic is completed autonomously by the initialization control logic unit (7) within the chip.

[0010] Specifically, the initialization control logic unit (7) also includes an enable terminal EN. The level state of the enable terminal determines whether initialization can be allowed. For example, in a live installation scenario, the magnetic field may not be determined due to position changes, and the magnetic field will only be fixed after the installation is completed. The initialization results during installation may be inaccurate, so as to avoid invalid initialization. In order to cooperate with power-on initialization, this port is enabled synchronously with power-on reset RESET when powered on.

[0011] In particular, when there is no need for I2C communication to inform the external processor of the specific magnetic field strength value, the I2C port is omitted and simplified to the initialization control logic unit (7) which only has the EN input terminal for initialization, the RESET input terminal for running initialization, and the GOOD signal input terminal for initialization completion, thus reducing costs.

[0012] The compensation and signal conditioning circuit unit (4) is a multi-stage analog signal processing link. Its input is connected to the differential output of the Hall semiconductor sensing unit (1), and its output is connected to the input of the analog-to-digital converter unit (5). Specifically, it includes the following sub-modules: Preamplifier (4.1): The weak differential voltage output by the Hall sensor unit is amplified in the first stage to suppress common-mode interference.

[0013] The instrumentation amplifier employs a low-noise, high common-mode rejection ratio (CMRR, typically >80dB) structure, with a gain typically fixed at 10 to 50 times. The input stage is usually a PMOS differential pair or a chopper-stabilized structure to reduce 1 / f noise.

[0014] Offset cancellation circuit (4.2): Eliminate the DC offset voltage present in the Hall sensor unit itself and the preamplifier to prevent the offset from being further amplified in subsequent amplification and affecting the initialization accuracy.

[0015] Chopper stabilization technology is employed: the offset voltage is modulated to a high frequency and then filtered out through input modulation, AC amplification, and demodulation. The chopper frequency is typically 10kHz to 500kHz. Alternatively, correlated double sampling is used: the signal and offset are sampled separately within the sampling period, and the difference is output. This reduces the equivalent input offset voltage to the μV level.

[0016] Temperature compensation bias circuit (4.3): It compensates for sensitivity drift and misalignment drift of the Hall sensor unit caused by temperature changes, so that the conditioned signal remains stable in a wide temperature range (such as -40℃ to 150℃).

[0017] An integrated on-chip temperature sensor monitors the chip temperature in real time. A programmable current / voltage source generates a temperature-proportional compensation bias, which is injected into the preamplifier or gain stage. The compensation parameters can be adjusted via an analog proportional adjustment circuit.

[0018] After temperature compensation, the sensitivity temperature coefficient of the Hall signal can be controlled within ±50ppm / ℃.

[0019] Programmable gain amplifier (4.4): The preamplified signal is further amplified to near the full scale of the analog-to-digital converter unit in order to make full use of the ADC's dynamic range.

[0020] It adopts a resistive feedback instrumentation amplifier, and the gain value is fixed before mass production, with a typical gain range of 1 to 32 times.

[0021] Low-pass filter (4.5): It filters out chopper switch noise, high-frequency electromagnetic interference, and out-of-band noise of the preamplifier to prevent noise from entering the analog-to-digital converter unit, while retaining the magnetic signal component.

[0022] It employs a first- or second-order active RC filter with a configurable cutoff frequency (e.g., 1kHz, 5kHz, 20kHz), but is fixed before mass production. The peak-to-peak noise of the output signal is reduced to tens of μV.

[0023] Output buffer (4.6): Provides low impedance drive capability for analog-to-digital converter unit (5) to ensure voltage stability of signal at the moment of sampling.

[0024] A unity-gain closed-loop operational amplifier with high input impedance, low output impedance (<100Ω) and sufficient slew rate (typically >1V / μs).

[0025] The cascading order of the sub-modules is as follows: Hall sensor unit (1) differential output → pre-differential amplifier (4.1) → offset cancellation circuit (4.2) → temperature compensation bias circuit (4.3) → programmable gain amplifier PGA (4.4) → low-pass filter LPF (4.5) → output buffer (4.6) → analog-to-digital converter unit (5). Among them, the temperature compensation bias circuit (4.3) can also provide bias current to the Hall sensor unit.

[0026] Collaboration with the initialization control logic unit of the analog-to-digital converter unit: In the initialization mode, the initialization control logic unit (7) configures the driving capability of the micro-coil drive circuit unit (3), including the driving voltage and driving current, and changes the direction and intensity of the applied magnetic field through the micro-coil circuit unit (2) to counteract the ambient magnetic field. After initialization is completed, the driving voltage and driving current are locked and the working mode is entered.

[0027] The initialization control logic unit (7) is configured to perform the following closed-loop operation in the initialization mode: 1. When EN is in the enabled state, the digital value of the Hall output voltage after conversion by the analog-to-digital converter unit is continuously acquired through the RESET signal. The RESET signal is generated by the RC hardware circuit when the circuit is powered on, and can also be generated by an external button or MCU at any time after power-on as needed. 2. Compare the current Hall output voltage value with a preset reference value to generate an error signal; wherein, the preset reference value can be zero (corresponding to a zero magnetic field target) or a programmable bias voltage configured by the user; 3. Output control quantity according to the error signal, which determines the direction and magnitude of the output current of the micro-coil drive circuit unit (3), thereby changing the reverse magnetic field strength generated by the micro-coil (2); 4. Repeat steps 1 to 3 above to form a closed-loop feedback control. In each iteration, the reverse magnetic field is adjusted in real time based on the latest Hall output until the Hall output voltage value is stably converged within the specified tolerance range of the preset reference value (e.g., the preset reference value is 0). 5. After step 4 above is completed, the analog-to-digital conversion unit generates a GOOD signal, then the initialization control logic unit latches and maintains the current micro-coil drive value, exits the initialization mode, and the sensor automatically enters the normal working state. After that, the Hall output signal is the compensated effective signal.

[0028] The initialization mode is triggered by any one or more of the following combinations: Automatic power-on trigger: Initialization is automatically initiated by the power-on reset signal, without any external intervention; External enable signal trigger: triggered by a level change of the dedicated enable pin RESET, or by an "on-demand initialization" command sent by an external MCU received through the I2C interface; Hall output over-limit event trigger: During normal operation, if the analog-to-digital converter unit initialization control logic unit detects that the Hall output voltage exceeds the preset threshold range (indicating a significant change in the ambient magnetic field), the initialization mode will be automatically re-triggered to achieve long-term adaptive calibration.

[0029] Key features: Throughout the initialization process (regardless of the triggering method), all operations involving error calculation, control law execution, and current drive adjustment are independently completed within the chip by the initialization control logic unit (7), without relying on any external microprocessor (MCU) for calculation, reading, or writing instructions. Even if the external MCU is connected to the I2C interface, it is only used to send trigger commands or read status values, without participating in the closed-loop control loop. This is fundamentally different from the open-loop scheme in the prior art (application number: 2024107114755, publication number: CN118501782A), where the external MCU must read the Hall value, calculate compensation parameters, and then write back instructions via I2C / SPI.

[0030] In a preferred embodiment, the initialization control logic unit (7) further includes a finite state machine (FSM) to manage different stages of the initialization mode: idle wait → trigger start → closed-loop adjustment → convergence determination → completion exit. The finite state machine is configured to execute proportional-integral control logic or successive approximation logic.

[0031] In this invention, the micro-coil circuit unit (2) and the Hall semiconductor sensing unit (1) are integrated together in the same chip or the same package to generate a reverse compensation magnetic field with controllable direction and adjustable intensity in the analog-to-digital converter unit. To balance magnetic field coupling efficiency, chip area and process compatibility, this invention provides the following three preferred integration schemes, which can be selected according to performance and cost requirements.

[0032] Option 1: CMOS process back-end metal spiral coil (on-chip integration) This solution is applicable to standard CMOS or BCD processes and requires no additional mask.

[0033] Structural Design: The micro-coil (2) is implemented as a planar spiral inductor using the back-end metal layers of the chip (such as M2, M3, M4). The coil shape is square, hexagonal, or circular, with its inner diameter aligned with or concentrically arranged with the sensitive area of ​​the Hall sensing unit (1). The typical number of turns is 4 to 16, and the line width and spacing are designed according to the metal layer thickness and current density requirements (e.g., line width 5 to 20 μm, spacing 5 to 10 μm).

[0034] Relative Position: The Hall sensing unit (1) is located on the bottom layer of the chip (such as a P-type epitaxial layer or an N-well), and the micro-coil (2) is arranged in the metal layer directly above the Hall unit, with the two vertically aligned. The magnetic field generated by the coil is perpendicular to the chip surface (i.e., the Z-axis direction), which matches the characteristic of most Hall elements to be sensitive to vertical magnetic fields.

[0035] Electrical connection: The two ends of the microcoil are connected to the underlying device through metal vias and led out to the output pad of the microcoil drive circuit unit (3). The drive circuit is also integrated on the same chip, and the output stage adopts an H-bridge structure to control the current direction.

[0036] Performance parameters: Under typical supply voltage of 3.3V or 5V, the coil resistance is about 20~100Ω, the drive current can reach 10~50mA, and the generated reverse magnetic field strength ranges from ±0.5mT to ±5mT (depending on the number of turns and current), which is sufficient to cancel out most motor background magnetic fields.

[0037] Advantages: Fully compatible with the process, no additional cost; high integration.

[0038] Limitations: Due to the limited thickness of the metal layer, the coil resistance is relatively high, resulting in high power consumption; the magnetic field strength is limited, making it suitable for moderate interference environments.

[0039] Option 2: Utilize a redistribution layer (RDL) or a thick metal layer coil (enhanced type). This solution is suitable for scenarios with higher requirements for reverse magnetic field strength (such as industrial strong magnetic environments), and can be formed into thick metal coils using the redistribution layer (RDL) in wafer-level packaging (WLP) process.

[0040] Structural Design: After the chip's front-end processes are completed, a thicker copper or gold metal layer (typically 5~20μm thick, much higher than the 0.5~1μm of standard CMOS metal layers) is deposited above the passivation layer, and a large-section spiral coil is formed by photolithography. The number of turns can be reduced to 2~8 turns, and the linewidth / spacing can be increased to 30~50μm, significantly reducing resistance (down to a few ohms).

[0041] Relative position: The RDL coil is also located directly above the Hall sensing unit, but at a slightly greater distance (there is a passivation layer in between), but the increased magnetic field efficiency brought by the thick metal can compensate for the distance loss.

[0042] Process compatibility: RDL is a standard wafer-level packaging step that requires no changes to the front-end processes, only an additional thick copper photolithography step in the back-end. It is suitable for Hall sensors using fan-in or fan-out packaging.

[0043] Performance parameters: coil resistance 5~15Ω, driving current up to 50~150mA, reverse magnetic field strength up to ±5mT~±20mT, meeting the requirements of strong interference scenarios.

[0044] Advantages: Low resistance, high magnetic field, and good heat dissipation.

[0045] Limitations: An additional RDL lithography layer is required, which slightly increases the cost compared to a pure CMOS solution, but is still far lower than that of discrete coils.

[0046] Option 3: Discrete microcoils within the package (hybrid integration) For special cases where existing Hall sensor chips cannot be modified in terms of metal wiring or where a very large reverse magnetic field is required, a discrete microcoil solution within the package can be used.

[0047] Structural Design: The Hall sensor unit (1) and the analog-to-digital converter unit initialization control logic and other circuits are integrated into a small-sized chip (die). A miniature air-core coil or magnetic core coil (such as a miniature planar coil or wire-wound micro-inductor made using PCB technology) is placed on the same package substrate or lead frame in a surface mount manner. The coil is connected to the drive circuit (still integrated into the main chip) through bonding wires or substrate wiring.

[0048] Relative position: The coil opening is directly opposite the sensitive area of ​​the Hall chip. The two are fixed by internal bonding or molding, and the distance between them is controlled within 0.1~0.5mm.

[0049] Process: Standard lead frame + injection molding packaging (such as SOP-8, QFN) is used, with an additional coil mounting step (which can be completed by the packaging factory).

[0050] Performance parameters: Coil resistance can be as low as 1~5Ω, drive current can reach more than 200mA, and reverse magnetic field strength can reach more than ±30mT, suitable for heavy motors or high current bus environments.

[0051] Advantages: High magnetic field strength, requiring no modifications to the original chip design.

[0052] Limitations: The package size increases slightly, and the mounting cost increases, but it is still a mature packaging technology.

[0053] Top Picks: For mid-to-high-end applications such as automotive electronics and industrial servo systems, Option 2 (RDL thick metal coil) is recommended because it achieves a good balance between performance, integration and cost, and supports large-scale mass production.

[0054] For low-cost consumer electronics or IoT devices, Option 1 (standard CMOS spiral coil) is recommended, requiring no additional process steps.

[0055] For special applications with extremely large background magnetic fields, Option 3 (discrete coils inside the package) can be used as a supplement.

[0056] Verification of magnetic field coupling between microcoil and Hall cell To ensure the effectiveness of closed-loop control, the magnetic flux density at the Hall sensing unit of the reverse magnetic field generated by the micro-coil should meet the following design criteria: At maximum drive current, the magnetic field strength generated by the micro-coil should be at least 1.2 times the expected maximum background magnetic field strength to ensure that the closed-loop system has sufficient adjustment margin.

[0057] The direction of the magnetic field must be strictly consistent with the sensitive axis of the Hall element (Z-axis for vertical Hall, X / Y-axis for horizontal Hall), and the coil layout must be aligned with the Hall element layout, with a deviation angle of less than ±5°.

[0058] Finite element simulation can be used to optimize the inner diameter, number of turns and spacing of the coil, so that the uniformity of the magnetic field in the Hall sensitive area is better than 90%, avoiding additional errors introduced by the magnetic field gradient.

[0059] Interface design between the drive circuit and the micro coil The interface between the micro-coil drive circuit unit (3) and the micro-coil (2) needs to consider overcurrent protection, reverse electromotive force absorption, and fast current switching. Specifically: The output stage of the drive circuit adopts an H-bridge structure, consisting of four power NMOS / PMOS transistors. The current amplitude is adjusted by PWM or linear control, and the magnetic field polarity is reversed by switching the direction of the H-bridge.

[0060] A freewheeling diode or active clamping circuit is connected in parallel across the micro-coil to absorb the reverse induced electromotive force generated at the moment of turn-off and protect the drive transistor.

[0061] The drive circuit integrates a current detection function (such as a series sampling resistor or a mirror current source) and feeds back the current value to the initialization control logic unit (7) for overcurrent protection and as a feedforward input for control.

[0062] Through the above system and method, this invention realizes a closed-loop on-chip system for the entire "perception-decision-execution" chain of Hall sensor initialization. While ensuring high accuracy and fast response, it greatly reduces the resource consumption of the system's main control unit and supports flexible external on-demand triggering. Beneficial effects

[0063] Compared with existing technologies (such as the solutions represented by application number: 2024107114755, publication number: CN118501782A), the present invention has the following beneficial effects: 1. Fully autonomous closed-loop initialization without external intervention: The entire process of "sensing (reading Hall value) - decision-making - execution (adjusting micro coil current)" is integrated on-chip, completely eliminating the dependence on external MCU and significantly reducing the software overhead and communication load of the system's main control. 2. Dynamic feedback control with high convergence accuracy: By monitoring the Hall output in real time and adjusting the reverse magnetic field in a closed loop, the Hall operating point can be accurately stabilized at a preset reference value (such as zero field), eliminating the residual error caused by "one-time calculation and fixed compensation" in open-loop control; 3. Fast response speed and strong real-time performance: No need for I2C / SPI communication round-trip delay, meeting the needs of scenarios with stringent initial response speed requirements; 4. Supports multiple trigger modes to enhance intelligence: In addition to automatic initialization upon power-on, it also supports external enable and overload initialization, enabling the sensor to have continuous adaptive capability and cope with sudden changes in the environmental magnetic field during operation. 5. Easy to industrialize and with obvious cost advantages: The closed-loop control logic only requires adding a small amount of digital logic resources to the existing Hall sensor chip, without adding external components or special processes, the increase in chip area is limited, and the cost is low; 6. Wide range of applications and great market potential: It is particularly suitable for low-cost IoT devices with limited MCU resources, automotive modules that require fast power-on response (such as EPS and electric water pumps), and Industry 4.0 self-calibrated sensors that need to meet functional safety requirements.

[0064] The aforementioned effects are not simply improvements on the prior application, but rather the creation of a completely new technical path of "on-chip autonomous closed-loop initialization," which has outstanding substantive features and significant progress.

[0065] Compared with the inventor's previously disclosed application 2024107114755, this invention establishes a completely new technical path from 'open-loop external dependence' to 'closed-loop on-chip autonomy', achieving fundamental improvements in control architecture, response speed, and system integration, and establishing a technical solution that is difficult to bypass through simple substitution.

[0066] This invention is a further improvement based on the applicant's previously disclosed application 2024107114755. By on-chiping the closed-loop control logic, a solution with greater technical depth and industrial application value is formed. Attached Figure Description

[0067] Appendix Figure 1 This is a diagram showing the system composition of the present invention. The main connections are as follows: 1. The output of the Hall semiconductor sensing unit (1) is connected to the input of the compensation and signal conditioning circuit unit (4) (differential signal); 2. The output of the compensation and signal conditioning circuit unit (4) is connected to the input of the analog-to-digital converter unit (5); 3. The digital output of the analog-to-digital converter unit (5) is connected to the Hall voltage digital value input of the initialization control logic unit (7); 4. The control output of the initialization control logic unit (7) is connected to the current direction and magnitude control input of the micro-coil drive circuit unit (3); 5. The drive output (H-bridge output) of the micro-coil drive circuit unit (3) is connected to both ends of the micro-coil circuit unit (2); 6. The power supply output of the power management unit (6) is connected to the power supply of all the above units (1, 2, 3, 4, 5, 7) respectively, and provides a power-on reset signal to the initialization control logic unit (7); 7. The initialization control logic unit (7) also communicates with the external controller (MCU) through its internal I2C interface (receive trigger command, send status flag), but this interface does not participate in closed-loop control.

[0068] The compensation and signal conditioning circuit unit (4) is a multi-stage analog signal processing link, and its internal sub-modules are connected in the following order: The differential output of the Hall semiconductor sensing unit (1) → the differential input of the preamplifier (4.1); The output of the preamplifier (4.1) → the input of the offset cancellation circuit (4.2); Output of offset cancellation circuit (4.2) → Signal input of temperature compensation bias circuit (4.3) (the bias injection point can be located at this stage or the subsequent PGA input; the temperature compensation bias circuit also provides additional bias current to the Hall sensing unit). The signal output terminal of the temperature compensation bias circuit (4.3) → the input terminal of the programmable gain amplifier (PGA, 4.4); Output of the programmable gain amplifier (4.4) → Input of the low-pass filter (LPF, 4.5); The output of the low-pass filter (4.5) → the input of the output buffer (4.6); The output of the output buffer (4.6) → the analog input of the analog-to-digital converter unit (5).

[0069] Closed-loop control related connections (signals and control): The analog-to-digital converter unit (5) sends the digitized Hall output voltage to the initialization control logic unit (7); the initialization control logic unit (7) calculates the control quantity (including current direction and magnitude information) and outputs it to the micro-coil drive circuit unit (3); the micro-coil drive circuit unit (3) drives the micro-coil circuit unit (2) to generate a reverse magnetic field according to the control quantity; the reverse magnetic field generated by the micro-coil circuit unit (2) is coupled to the Hall semiconductor sensing unit (1) to form a closed-loop feedback path.

[0070] The micro-coil drive circuit unit (3) integrates a current detection function to feed back the detected micro-coil current value to the initialization control logic unit (7) (for overcurrent protection and feedforward control).

[0071] External interactive connections: The I2C interface of the initialization control logic unit (7) is connected to the I2C bus of the external MCU through the chip pin to receive the "initialize on demand" command and send the initialization status flag; the external enable signal EN (if there is a dedicated pin) is connected to the trigger input of the initialization control logic unit (7); the Hall output over-limit event signal (generated by the analog-to-digital converter unit (5)) is connected to the event trigger input of the initialization control logic unit (7). The Hall sensor adaptive initialization system based on closed-loop feedback proposed in this invention has clear industrial manufacturing feasibility and broad application prospects.

[0072] Manufacturing feasibility: All constituent modules (Hall sensing unit, microcoil, ADC, digital control logic, etc.) can be implemented on standard CMOS or BCD integrated circuit processes. The microcoil can be formed by on-chip metal spiral inductors or by back-end processes (such as RDL layers). The control logic is a typical digital circuit, which is easy to synthesize and place. The initialization control logic unit (7) can be directly integrated into the digital back end of existing Hall sensor products without modifying the main structure of the analog front end.

[0073] Typical application scenarios: Common scenarios include applications where the environment around the device becomes magnetized after prolonged use. Another example is in electric vehicle motor controllers, where Hall sensors are installed at the motor end or near the busbar, where the background magnetic field is strong and dynamically changing. With this invention, the sensor automatically completes reverse magnetic field compensation within 30ms of power-on. The main control MCU does not need to periodically read and write compensation parameters, thus freeing up communication bandwidth (such as CAN / LIN) and avoiding untimely initialization issues caused by MCU software delays, thereby improving system real-time performance and ASIL functional safety level.

[0074] Industrial servo drive scenario: In a multi-axis system, each axis is independently equipped with a Hall current sensor. This invention enables each sensor to initialize autonomously upon power-up, eliminating the need for individual configuration by a host computer and simplifying the system power-up process.

[0075] In low-cost IoT magnetic switch / position sensor scenarios, where MCU resources are extremely limited (e.g., 8-bit microcontrollers with only a few hundred bytes of RAM), complex compensation algorithms cannot be run. This invention pushes all intelligent logic down to the sensor itself, requiring the MCU to simply read the initialization completion flag or directly use the Hall output, significantly reducing the demands on the main controller.

[0076] Industrialization advantages: Based on the existing published invention (2024107114755), this invention can achieve functional upgrades simply by adding on-chip closed-loop control logic. It does not require changes to the package pins or external application circuits, and can quickly replace traditional open-loop programmable Hall sensor products, forming a new type of Hall IC with independent high added value and strong market competitiveness. Detailed Implementation

[0077] The technical solution of the present invention will be described in detail below with reference to a preferred embodiment. This embodiment takes automatic power-on trigger initialization as an example to realize closed-loop regulation.

[0078] System connection and initial state: After the Hall sensor integrated circuit is powered on, the power management unit (6) generates a stable internal power supply and a power-on reset signal. This reset signal directly triggers the initialization control logic unit (7) to enter the initialization mode. At this time, the initial output of the micro-coil drive circuit unit (3) is zero, and the micro-coil circuit unit (2) does not generate a reverse magnetic field.

[0079] Closed-loop initialization process: 1. Signal Acquisition: The Hall semiconductor sensing unit (1) senses the external environmental magnetic field (e.g., leakage magnetic field from the motor rotor) and outputs a differential voltage. After being amplified and filtered by the compensation and signal conditioning circuit unit (4), the voltage is converted into a digital quantity with 12-bit resolution by the analog-to-digital converter unit (5) and sent to the initialization control logic unit (7). 2. Error Calculation: The internal register of the initialization control logic unit (7) stores a preset reference value. In this embodiment, to eliminate the influence of the background magnetic field on the Hall operating point, the preset reference value is set to zero (corresponding to the ideal output of zero magnetic field). The logic unit subtracts the reference value (0) from the current Hall output voltage digital value (e.g., corresponding to a +15mV equivalent magnetic field) to obtain the error signal value; 3. Dynamic Adjustment: The micro-coil drive circuit unit (3) drives the micro-coil (2) according to the control quantity to generate a reverse compensation magnetic field. This reverse magnetic field acts on the Hall sensing unit (1), causing the Hall output voltage to decrease. In the next sampling cycle, the analog-to-digital converter unit (5) reads a new, smaller output voltage value (e.g., +3mV), the error decreases, and the controller correspondingly reduces the output current. After several cycles (approximately 25ms in this embodiment, approximately 500 control cycles), the Hall output voltage enters the preset reference value ± tolerance range (e.g., the tolerance is set to ±0.5% of the full scale, corresponding to ±1mV). At this time, the error signal approaches zero, and the control quantity stabilizes at a non-zero final value (this value just generates the reverse magnetic field required to cancel the background magnetic field). 4. Initialization complete: After the initialization control logic unit (7) monitors the GOOD signal and confirms that the initialization convergence is complete for a specified time (e.g., 5ms), it determines that the initialization convergence is complete. The logic unit sets the internal "initialization complete" flag, the initialization control logic unit locks the control value, and exits the initialization mode. The sensor then enters normal working state. After this, the Hall output signal is the compensated linear signal that reflects the magnetic field of the target to be measured.

[0080] In summary, this invention, with on-chip integrated closed-loop feedback as its core, achieves efficient, autonomous, and accurate initialization of Hall sensors, possessing outstanding substantive features and significant progress.

Claims

1. A Hall sensor adaptive initialization system based on closed-loop feedback, integrated within a Hall sensor integrated circuit, characterized in that, include: Hall semiconductor sensing unit (1) is used to sense external magnetic fields and output electrical signals; The micro-coil circuit unit (2) is integrated inside the chip and is used to generate a reverse compensation magnetic field; The micro-coil drive circuit unit (3) is used to adjust the direction and magnitude of the current flowing through the micro-coil; The compensation and signal conditioning circuit unit (4) conditions the signal output by the Hall sensor unit; The analog-to-digital converter unit (5) converts the conditioned analog signal into a digital signal, and has an integrated window comparator. The power management unit (6) supplies power to each module and generates a power-on reset signal; An initialization control logic unit (7) is connected to the analog-to-digital converter unit (5) and the micro-coil drive circuit unit (3), respectively. It contains a finite state machine (FSM), which is configured to execute proportional-integral control logic or successive approximation logic. The initialization control logic unit (7) is configured to operate in initialization mode: The system continuously acquires the digital value of the Hall output voltage, compares it with a preset reference value to generate an error signal, and dynamically adjusts the direction and magnitude of the micro-coil current according to the error signal through the finite state machine (FSM). This process is repeated until the window comparator outputs a GOOD signal, and then exits the initialization mode.

2. The system according to claim 1, characterized in that, The initialization mode is triggered by any one or more of the following methods: automatic power-on triggering, external trigger signal triggering, Hall output over-limit event triggering; the initialization control logic unit (7) also includes an enable terminal, the level of which determines whether initialization is allowed.

3. The system according to claim 1, characterized in that, The initialization control logic unit (7) also includes an initialization control port, which is implemented using an I2C interface and is used to interact with an external controller to receive on-demand initialization trigger commands and / or provide initialization status flags to the outside. The I2C interface does not participate in parameter calculation or current regulation command transmission in the closed-loop feedback control process.

4. The system according to claim 1, characterized in that, The micro-coil circuit unit (2) and the Hall semiconductor sensing unit (1) are integrated in the same chip or in the same package, and the integration scheme is selected from any of the following: Option 1: Metal spiral coil in the back-end of CMOS process; Option 2: Utilize rewire layers or thick metal layer coils; Option 3: Discrete micro-coils within the package.

5. The system according to claim 1, characterized in that, The compensation and signal conditioning circuit unit (4) includes a pre-differential amplifier (4.1), an offset cancellation circuit (4.2), a temperature compensation bias circuit (4.3), a programmable gain amplifier (4.4), a low-pass filter (4.5), and an output buffer (4.6) cascaded in sequence; wherein the temperature compensation bias circuit (4.3) integrates an on-chip temperature sensor and generates a compensation bias that is proportional to or complementary to the temperature.

6. The system according to claim 1, characterized in that, The output stage of the micro-coil drive circuit unit (3) adopts an H-bridge structure. The micro-coil is connected in parallel with a freewheeling diode or an active clamping circuit. The drive circuit integrates a current detection function and feeds back the current value to the initialization control logic unit (7).

7. An adaptive initialization method for a Hall sensor based on closed-loop feedback, applied to the system described in any one of claims 1 to 6, characterized in that, Includes the following steps: Step A: Enter initialization mode and initialize the control logic unit (7) to continuously acquire the digital value of the Hall output voltage after analog-to-digital conversion; Step B: Compare the current Hall output voltage value with the preset reference value to generate an error signal; Step C: Based on the error signal, the proportional-integral control logic or successive approximation logic is executed through a finite state machine (FSM) to dynamically calculate the control quantity. The control quantity determines the direction and magnitude of the output current of the micro-coil drive circuit unit (3), thereby changing the reverse magnetic field strength generated by the micro-coil. Step D: Repeat steps A to C to form a closed-loop feedback control until the window comparator output by the analog-to-digital converter unit generates a GOOD signal; Step E: Exit initialization mode, and the sensor enters normal working state.

8. The method according to claim 7, characterized in that, The initialization mode is triggered by any one or more of the following methods: automatic triggering by power-on reset signal, triggering by external enable signal, or triggering by Hall output over-limit event.

Images