Picoampere-level current measuring device with leakage estimation and digital compensation and picoampere-level current measuring method

Abstract

The invention provides a picoampere-level current measuring device and method with leakage estimation and digital compensation, and the device comprises an electrometer-level operational amplifier and a transimpedance feedback network which are used for converting a measured current into a voltage signal; in combination with a retaining ring conductor driven by a retaining ring buffer and an inner and outer double-layer metal shielding structure, high-resistance node leakage and external interference are remarkably reduced. The digital controller injects 5mV-50mV micro excitation into the retaining ring in a non-sampling window and detects output response so as to estimate input equivalent leakage conductance and perform online compensation; and meanwhile, environment self-adaptive correction is carried out according to a temperature and humidity-leakage admittance model. The system adopts a 24-bit analog-to-digital converter, two-stage zero-drift amplification and multi-point linearity calibration can be switched, picoampere-level current measurement with high precision, low noise and wide dynamic range is realized, and the system is suitable for the fields of high-insulation detection, light current measurement, electrochemical micro-current analysis and the like.

Description

Technical Field

[0001] This invention relates to the field of current measurement technology, and specifically to a picoampere-level current measurement device and method with leakage estimation and digital compensation. Background Technology

[0002] Picoampere-level current measurement is widely used in fields such as photoelectric detection, ion current detection, electrochemical ultra-high resistance sensor readout, and leakage current testing of high insulation materials.

[0003] The shortcomings of existing technology: 1. The coupling between high-resistivity nodes and the board surface is difficult to completely eliminate. Even with guard rings, traditional planar wiring inevitably results in surface conductive coupling between high-resistance nodes at the inverting end and the FR-4 board surface, residual flux on the pads, silkscreen printing, and the dust-absorbing layer. Due to manufacturability and size constraints, creepage distances are often insufficient, leading to significant fluctuations in leakage admittance in high-humidity environments or during temperature drift, making it difficult to establish a stable error model.

[0004] 2. Retaining ring drives are mostly static, lacking observability of leaks. Most existing retaining rings use constant equipotential drive, primarily serving an "isolation / shielding" function, but they cannot directly quantify the equivalent leakage conductance of the current device under specific environmental conditions. In engineering, they typically rely on offline calibration or periodic no-load measurements, which are insufficient to reflect leakage changes caused by factors such as increased humidity, contamination accumulation, and connector replacement, leading to a lack of or delay in online compensation.

[0005] 3. Calibration and compensation lack online adaptive capabilities. Traditional multi-point linearity calibration is mostly performed in a laboratory environment. However, changes in temperature, humidity, and insulation conditions at the application site can cause the calibration to deviate from the actual operating point. Without environmental awareness and adaptive update mechanisms, zero-point drift, scale factor drift, and leakage errors may accumulate, ultimately limiting long-term stability and traceability.

[0006] 4. Incompatible range management and overload recovery. Most devices employ a fixed post-stage gain or a simple range switching strategy. When the measured signal abruptly approaches saturation, the loop recovery and reading linearization are poor, which affects the short-time dynamic response and may lead to additional error accumulation.

[0007] Therefore, existing technologies have shortcomings and need further improvement. Summary of the Invention

[0008] To address the problems existing in the prior art, the present invention provides a picoampere-level current measurement device and method with leakage estimation and digital compensation.

[0009] To achieve the above objectives, the specific solution of the present invention is as follows: The application provides a picoampere current measurement device with leakage estimation and digital compensation, comprising: An operational amplifier and a transimpedance feedback network thereof are used to convert the measured current into voltage; The feedback network comprises a feedback resistor of 1GΩ-100GΩ and a parallel compensation capacitor; The inverting input of the operational amplifier and the feedback network are connected by a column suspension to be suspended in the air, so that the high-resistance node of the inverting terminal is not in direct contact with the FR-4 board surface; A guard ring conductor driven by a guard ring buffer is arranged around the high-resistance node of the inverting terminal; The device adopts a double-layer metal shielding structure, wherein the inner layer shielding is connected to the analog ground, and the outer layer shielding is connected to the chassis ground; a shielding connector is arranged at the input end, and three coaxial interfaces are supported; The current measurement device further comprises an environment sensing module, an analog-to-digital converter and a digital controller, wherein the digital controller injects a micro-excitation with an amplitude of 5mV-50mV into the guard ring conductor in a non-sampling window and measures the transimpedance output response to estimate the input equivalent leakage conductance and compensate the measured current online.

[0010] Further, the compensation capacitor is 1pF-10pF to ensure the phase margin of the transimpedance amplifier, and an RC low-pass filter is arranged at the output end to suppress high-frequency noise and out-of-band interference.

[0011] Further, the column suspension connection comprises a rigid column and an insulating support; the inverting terminal pin is directly welded to the suspended end of the feedback resistor / capacitor integrated component through a short lead, and the integrated component is fixed by the insulating support, so that the shortest creepage path of the inverting node to the board surface is greater than 8mm.

[0012] Further, the micro-excitation is controlled by a programmable micro-step potential generator of the digital controller and added to the guard ring conductor through the guard ring buffer; the micro-excitation step duration is 0.5ms-5ms, and the sampling of the measured input is paused during the duration; the digital controller estimates the equivalent leakage conductance according to the first-order or second-order difference of the output response, and updates the compensation parameters with an adaptive refresh period T=1s-60s.

[0013] Further, an insulating gasket is arranged between the inner layer and the outer layer of the double-layer metal shielding structure; the outer layer shielding is directly connected to the chassis ground, and the inner layer shielding is coupled to the chassis ground at a single point through a capacitor or a resistor-capacitor network to suppress circulating current; the shielding box gap adopts a stepped lap joint to reduce the gap coupling.

[0014] Further, the input end can be configured in two modes of an SMA interface or a triaxial interface; when the SMA interface is selected, the SMA outer conductor is made equipotential with the high-impedance node under the low-output impedance drive of the guard ring buffer through a special guard ring adapter, and is connected to the chassis ground through an independent outer shield, so as to reduce the cable leakage and maintain the chassis shielding integrity.

[0015] Further, a switchable second-stage zero-drift amplifier and an analog switch array are connected in series after the transimpedance output, so as to realize the selection of the second-stage gain G2=1, 2, 5, 10; and the second-stage amplification is automatically bypassed when the output approaches saturation, so as to expand the linear range.

[0016] Further, the environmental perception module comprises temperature and humidity sensors; and the digital controller performs temperature and humidity compensation on the zero point and the scale factor based on a temperature and humidity-leakage admittance empirical model.

[0017] Further, the analog-to-digital converter is 24-bit and has a sampling rate of ≥2k SPS; and in combination with segmented multi-point (not less than 80 points) linearity calibration and a programmable integration window, the low-frequency 1 / f noise is suppressed by using a sliding average or window integration.

[0018] The application further provides a pico-ampere level current measurement method with leakage estimation and digital compensation, applied to the above device, and the method comprises the following steps: S1, setting a transimpedance feedback parameter and a compensation capacitance value; S2, collecting a transimpedance output voltage and converting a current according to a calibration coefficient; S3, injecting a micro-excitation into the guard ring conductor in a non-sampling window and measuring an output response, and pausing the sampling of the measured input during the micro-excitation, so as to estimate an equivalent leakage conductance of the input; S4, performing online compensation on current data according to the leakage estimation and a temperature and humidity model; S5, adaptively selecting a second-stage gain and an integration window according to an output amplitude, so as to balance the resolution and the dynamic range.

[0019] The technical scheme of the application has the following beneficial effects: 1. Significantly reducing the leakage of the high-impedance node and the surface adsorption current By adopting the column suspension connection structure, the inverting terminal of the operational amplifier is directly connected with the feedback network in the air, so that the high-impedance node is no longer in direct contact with the FR-4 board surface; in combination with the mechanical support design with a creepage distance greater than 8 mm, the leakage channel caused by the board surface moisture absorption, pollution or silk printing residue is effectively avoided, the equivalent leakage conductance is significantly reduced, and the long-term stability is improved.

[0020] 2. Realizing real-time estimation and online digital compensation of the leakage admittance The traditional galvanometer can only correct errors by static calibration. The present application injects a micro-excitation signal of 5mV-50mV into the guard ring conductor in a non-sampling window, measures the trans-impedance output response, and uses a first-order or second-order difference algorithm of a digital controller to estimate the input equivalent leakage conductance in real time. Thus, online dynamic compensation of leakage errors can be realized, and high-precision measurement can be maintained under conditions of temperature and humidity or environmental changes.

[0021] 3. Transition of the guard ring from static isolation to measurable and controllable dynamic guard ring The present application combines guard ring driving with micro-excitation functions, so that the guard ring not only maintains high-resistance node equipotential, but also has active excitation and response detection functions, realizing self-checking of the guard ring state and admittance feedback. This design improves the self-adaptive ability of the system to environmental disturbances, and transforms the guard ring from a passive protection element to a tunable and observable functional module.

[0022] 4. Construction of a multi-layer shielding system to eliminate loop current and gap coupling Through a double-layer metal shielding structure, the inner layer shielding is connected to the analog ground, the outer layer shielding is connected to the chassis ground, and they are coupled at a single point by an RC network, which can effectively prevent potential drift between the ground loop current and different grounding points. The stepped lap joint of the shielding box gap greatly reduces the gap coupling effect of the external electromagnetic field, thereby significantly improving the anti-interference ability and electromagnetic compatibility performance of the entire machine.

[0023] 5. Compatibility with SMA and triaxial interfaces, effectively suppressing cable leakage The device input end is compatible with SMA and triaxial interfaces. When SMA is selected, the outer conductor is driven to equipotential by the guard ring buffer, and the independent outer shielding is connected to ground, realizing a "driven shielding + chassis shielding" layered structure, which not only maintains the interface universality, but also significantly reduces the leakage and noise coupling caused by the connector and cable medium.

[0024] 6. Self-adaptive gain and wide dynamic range characteristics A switchable two-stage zero-drift amplifier and an analog switch array are set after the trans-impedance output, realizing multi-stage gain selection of G2=1, 2, 5, and 10. When the output approaches saturation, the system automatically bypasses the second-stage amplifier, thereby automatically expanding the dynamic range and preventing output saturation distortion, realizing the compatibility of weak signal measurement and large dynamic range signal processing.

[0025] 7. Environment perception and temperature and humidity self-compensation mechanism Through real-time monitoring of environmental changes by temperature and humidity sensors, and combined with the temperature and humidity-leakage admittance empirical model in the digital controller for adaptive correction of the zero point and the proportional factor, the drift caused by environmental changes can be effectively eliminated, and the long-term stability and repeatability of the measurement results can be improved.

[0026] 8. Coexistence of high resolution and low noise The 24-bit high-precision analog-to-digital converter (≥2 kSPS) is combined with segmented multi-point (≥80 points) linearity calibration, and the programmable integral window and the sliding average algorithm are combined, so that the low-frequency 1 / f noise and the out-of-band interference can be obviously inhibited, and the high-resolution output of the pA or even the fA level current measurement can be realized.

[0027] 9. Dynamic parameter refresh and self-correction mechanism The system updates the compensation parameters with an adaptive refresh period T=1s~60s, so that the compensation model can automatically adjust with the environment and time, the high precision and high stability of the system in long-term operation are maintained, and the tedious operation of manual recalibration is avoided. BRIEF DESCRIPTION OF DRAWINGS

[0028] Fig. 1 It is the principle block diagram of the application; Fig. 2 It is the overall flowchart of the application; DETAILED DESCRIPTION The application will be further described below in combination with the drawings and embodiments; it can be understood that the specific embodiments described herein are only used to explain the application, rather than limit the application; in addition, it should be noted that, in order to facilitate the description, only the parts related to the application are shown in the drawings, rather than all.

[0029] In combination with Figs. 1-2 As shown in the figure, the application provides a pA level current measurement device with leakage estimation and digital compensation, comprising: A pA level operational amplifier and its transimpedance feedback network are used to convert the measured current into voltage; The feedback network contains a feedback resistor of 1GΩ~100GΩ and a parallel compensation capacitor; The inverting input end of the operational amplifier and the feedback network are connected by a column suspension to be suspended in the air, so that the high-resistance node of the inverting end is not in direct contact with the FR-4 board surface; A guard ring conductor driven by a guard ring buffer is arranged around the high-resistance node of the inverting end; The device adopts an inner-outer double-layer metal shielding structure, wherein the inner layer shielding is connected to the analog ground, and the outer layer shielding is connected to the chassis ground; a shielding connector is provided at the input end, and three coaxial interfaces are supported; The current measurement device further comprises an environment perception module, an analog-to-digital converter and a digital controller, the digital controller injects a micro-excitation with an amplitude of 5mV~50mV into the guard ring conductor in a non-sampling window and measures the transimpedance output response to estimate the input equivalent leakage conductance and compensate the measured current online.

[0030] The compensation capacitor is 1pF~10pF to ensure the phase margin of the transimpedance amplifier, and an RC low-pass filter is configured at the output end to suppress high-frequency noise and out-of-band interference.

[0031] The stand column suspension connection comprises a rigid stand column and an insulating support; the reverse phase end pin is directly welded to the suspension end of the feedback resistance / capacitance integrated element through a short lead line, and the integrated element is fixed through the insulating support, so that the shortest creepage path from the reverse phase node to the board surface is greater than 8 mm.

[0032] The micro-excitation is controlled by a digital controller, is added to the guard ring conductor through a guard ring buffer, and has a micro-excitation step duration of 0.5 ms to 5 ms, and during the duration, sampling of the measured input is suspended; the digital controller estimates the equivalent leakage conductance according to the first-order or second-order differential of the output response, and updates the compensation parameters with an adaptive refresh period T = 1 s to 60 s.

[0033] An insulating gasket is arranged between the inner layer and the outer layer of the double-layer metal shielding structure; the outer layer shielding is directly connected with the case ground, and the inner layer shielding is coupled with the case ground at a single point through a capacitor or a resistance-capacitance network to suppress circulating current; the shielding box gap adopts a stepped lap joint to reduce the gap coupling.

[0034] The input end has two configurations of an optional SMA interface or a three-coaxial interface; when the SMA interface is selected, the SMA outer conductor is made equal to the high resistance node under the low output impedance driving of the guard ring buffer by means of a special guard ring adapter, and is connected with the case ground through an independent outer layer shielding to reduce cable leakage and maintain the integrity of the case shielding.

[0035] A switchable two-stage zero drift amplifier and an analog switch array are connected in series after the transimpedance output to realize the selection of a two-stage gain G2 = 1, 2, 5, 10; when the output approaches saturation, the two-stage amplification is automatically bypassed to expand the linear range.

[0036] The environmental perception module comprises temperature and humidity sensors; the digital controller performs temperature and humidity compensation on the zero point and the proportional factor based on a temperature and humidity-leakage admittance empirical model.

[0037] The analog-to-digital converter is 24-bit and has a sampling rate ≥ 2k SPS; in combination with segmented multi-point (not less than 80 points) linearity calibration and a programmable integration window, low-frequency 1 / f noise is suppressed by using sliding average or window integration.

[0038] The application also provides a picoampere-level current measurement method with leakage estimation and digital compensation, which is applied to the device and comprises the following steps: S1, setting transimpedance feedback parameters and compensation capacitance values; S2, collecting a transimpedance output voltage and converting a current according to a calibration coefficient; S3, injecting a micro-excitation into a guard ring conductor and measuring an output response in a non-sampling window, and suspending sampling of a measured input during the micro-excitation to estimate an input equivalent leakage conductance. S4, compensating the current data on-line according to the leakage estimation and the temperature and humidity model; S5, selecting the secondary gain and the integration window adaptively according to the output amplitude, so as to balance the resolution and the dynamic range.

[0039] I. General principle The present application belongs to the technical field of picoampere-level current precision measurement, and the core idea is to introduce active leakage estimation and digital compensation mechanism on the basis of high-resistance trans-impedance amplification structure.

[0040] The device takes a picoampere-level current as the core, converts the picoampere-level current into a voltage signal, and processes the voltage signal by a digital controller after secondary amplification and analog-digital conversion. The system simultaneously injects a micro-excitation signal into the guard ring conductor in a non-sampling window, measures the transient response of the trans-impedance output, calculates the equivalent leakage conductance of the high-resistance node, and then uses the result to correct the actual measurement current in real time.

[0041] The whole system forms a closed-loop measurement system integrating high insulation structure, guard ring active detection, digital compensation algorithm and environmental adaptive model, realizes dynamic perception and automatic correction of leakage error, and thus maintains high measurement precision and stability in long-term operation and complex environment.

[0042] II. Trans-impedance amplification and high-resistance node structure The measured current enters the trans-impedance amplifier module through the shielded connector. The inverting input of the picoampere-level operational amplifier is connected to the feedback network in the air through a column suspension connection, avoiding direct contact between the high-resistance node and the FR-4 board. The feedback network is composed of 1GΩ-100GΩ feedback resistor and 1pF-10pF compensation capacitor, which is used to set the trans-impedance gain and stabilize the loop phase margin.

[0043] Due to the high feedback resistance, any weak leakage or parasitic capacitance current will introduce errors. Therefore, a guard ring conductor is arranged around the inverting node, which is driven by a low-output-impedance guard ring buffer, so that the guard ring and the high-resistance node are at the same potential, and the surface current and parasitic capacitance coupling are suppressed. The column suspension structure and the guard ring together form a three-dimensional high insulation isolation area, effectively reducing the leakage channel and displacement current coupling.

[0044] III. Guard ring micro-excitation and leakage estimation principle In order to obtain the current leakage status of the high-resistance node, the digital controller injects a micro-excitation signal with an amplitude of 5mV-50mV into the guard ring conductor through the guard ring buffer by a programmable micro-step potential generator in a non-sampling window period, and the excitation duration is 0.5ms-5ms.

[0045] The micro-excitation will cause a small response voltage in the feedback path of the system, reflecting the change of the admittance of the equivalent leakage channel between the guard ring and the high resistance node. After the analog-to-digital converter records the transimpedance output response, the controller calculates the first or second order difference, from which the estimated value of the equivalent leakage conductance is extracted.

[0046] Through periodic excitation and detection, dynamic monitoring of the leakage channel admittance can be achieved, and a real-time leakage estimate G_leak(t) is obtained.

[0047] Four, digital compensation algorithm and adaptive refresh The digital controller combines the estimated G_leak(t) with the transimpedance gain parameter Rf to calculate the leakage current component I_leak = Vout × G_leak(t), and real-time deducts it from the measured current, thereby completing online digital compensation.

[0048] The compensation parameter is not a fixed value, but is adaptively refreshed according to environmental and time changes. The controller re-executes the micro-excitation estimation and parameter update with a period of 1s-60s to form a closed-loop correction mechanism, so that the system remains in a stable calibration state under conditions of temperature and humidity changes, pollution accumulation or mechanical stress changes.

[0049] Five, signal acquisition and noise suppression path The voltage signal after transimpedance amplification enters the switchable two-stage zero drift amplifier and analog switch array, realizing multiple gain G2=1, 2, 5, 10. When the output approaches saturation, the system automatically bypasses the second-stage amplifier to expand the linear range.

[0050] The signal then passes through an RC low-pass filter to suppress high-frequency noise and out-of-band interference, and then enters a 24-bit high-resolution analog-to-digital converter. The analog-to-digital converter uses a sampling rate of ≥2 kSPS, combined with segmented multi-point linearity calibration, programmable integration window and sliding average algorithm, effectively reducing 1 / f noise and quantization error, ensuring the linearity and stability of the output signal.

[0051] Six, shielding, grounding and interface coupling control The device uses an inner and outer double-layer metal shielding structure. The inner layer shields the analog ground, used to isolate the amplifier and high resistance node area; the outer layer shields the chassis ground, and is coupled to the inner layer at a single point through an RC network to suppress ground loop and power frequency interference.

[0052] The input end can be configured with SMA or triaxial interface. When SMA is selected, the outer conductor of the SMA is driven to be at the same potential as the high resistance node through the guard ring buffer, and the outer shield is independently connected to the chassis ground, realizing "inner drive and outer shield" layered shielding, effectively eliminating cable medium leakage and external electromagnetic coupling.

[0053] Seven, environmental perception and temperature and humidity compensation mechanism The environmental sensing module is provided with a temperature and humidity sensor, which collects environmental parameters in real time and sends them to a digital controller. The controller corrects zero drift and proportional coefficient according to the temperature and humidity-leakage admittance empirical model, and optimizes the digital filter window length and refresh cycle.

[0054] When the material admittance deviates due to the increase of environmental humidity or temperature change, the controller can synchronously adjust the leakage estimation model and compensation parameters to maintain the accuracy of the output reading.

[0055] Eight, overall operation process After the system is powered on, the controller initializes the transimpedance feedback and filter parameters; then starts the normal sampling stage (S2) and intermittently enters the window period (S3). In each window period, the controller performs the guard ring micro-excitation and response sampling, calculates the leakage admittance, and updates the compensation coefficient; the current value measured in the next sampling period is automatically deducted from the leakage component, which is output after two-stage amplification, analog-to-digital conversion, and digital filtering.

[0056] The whole process forms a continuous "measurement-estimation-compensation-update" cycle, realizing dynamic correction of high resistance measurement errors. Through structural and algorithmic means, the present application can realize high-precision, wide-dynamic, and long-term stable automatic measurement in the range of picoampere or even nanoampere current.

[0057] Embodiment 1: Picoampere current measurement device with leakage estimation and digital compensation One, purpose and scope of application of the embodiment The present embodiment gives a set of reproducible engineering solutions for the "picoampere current measurement device with leakage estimation and digital compensation" described in the claims, The content covers structural design, key parameters, component selection, mechanical assembly, shielding and grounding, firmware and algorithmic process, calibration and test method, and operation and maintenance specification.

[0058] The present embodiment strictly uses the terms and module names consistent with the claims, including: an electrical metering operational amplifier, a transimpedance feedback network (1GΩ-100GΩ in parallel with 1pF-10pF compensation capacitor), a column suspension connection (in-air suspension direct connection), a guard ring buffer and a guard ring conductor, an inner and outer double-layer metal shielding structure (the inner layer shielding is connected to the analog ground, and the outer layer shielding is connected to the chassis ground), a shielded connector (supporting SMA / three coaxial interface), an environmental sensing module, an analog-to-digital converter and a digital controller (injecting a micro-excitation of 5mV-50mV into the guard ring conductor and collecting the response to estimate the input equivalent leakage conductance and compensate online in the non-sampling window), an RC low-pass filter, a two-stage zero-drift amplifier, and an analog switch array. The application objects include pA-nA order measurements of photoelectric current, material insulation leakage, electrochemical micro-current, and ion current.

[0059] II. Structure Overview and Connection Relationship The signal and control path of the device from input to output is in sequence: input end (shielded connector, supporting SMA or triaxial interface) → operational amplifier and transimpedance feedback network (Rf=1GΩ~100GΩ in parallel with Cf=1pF~10pF) → RC low-pass filter → two-stage zero-drift amplifier → analog switch array (realizing G2=1, 2, 5, 10 selection and automatic bypass) → analog-to-digital converter (24-bit, ≥2kSPS) → digital controller.

[0060] The digital controller is connected with the environment sensing module at the same time, and outputs a micro-excitation through a programmable micro-step potential generator, which is added to the guard ring conductor through a guard ring buffer; The guard ring conductor is arranged around the high-resistance node of the inverting terminal, and the inverting terminal and the feedback network are connected in the air through a stand suspension, ensuring that the shortest creepage distance is ≥8mm.

[0061] The external shielding adopts an inner-outer double-layer metal shielding structure: the inner layer shielding is connected to the analog ground, and the outer layer shielding is connected to the chassis ground, and the two are coupled at a single point through a capacitor or a resistance-capacitance network to suppress circulating current; The shielding box gap adopts a stepped lap joint to reduce the gap coupling.

[0062] III. Target Index and Design Boundary (1) Measurement range: according to the selection of the transimpedance feedback network, typically set to four levels: Rf=1GΩ, 10GΩ, 50GΩ, 100GΩ.

[0063] (2) Resolution: under the conditions of Rf=100GΩ, integration time 1s, and triaxial input, the effective resolution is better than 5fA_rms.

[0064] (3) Zero point drift: 25 °C±2 °C, 40%RH±5%RH, continuous 24 h test, zero point drift not more than 50 fA (compensated by digital).

[0065] (4) Frequency response: RC low-pass cutoff is about 160 Hz (R=1 kΩ, C=1 µF), which meets the basic anti-aliasing before ≥2 kSPS sampling.

[0066] (5) Stability: no self-excitation oscillation, no saturation locking after long-time work (>1000 h); the guard ring micro-excitation refresh period T can be set to 1~60 s (typical 5 s).

[0067] (6) Environment: 10~70%RH, -10~55 °C; it is recommended to be equipped with a desiccant and anti-condensation measures in a humid heat environment.

[0068] The above indexes are typical values that can be achieved by the embodiment, and the specific values can be adjusted within the range to match different measured objects and cable capacitances.

[0069] IV. Detailed design and parameters of each module 4.1 Electrically-cascaded operational amplifier and transimpedance feedback network The electrically-cascaded operational amplifier is selected from a FET input type device with an input bias current ≤ 100 fA (25 °C, typical), an input impedance ≥ 10^14 Ω, and an input voltage noise density ≤ 5 nV / √Hz (> 10 Hz) at a supply voltage of ±5 V.

[0070] The transimpedance feedback network is composed of a feedback resistor Rf and a parallel compensation capacitor Cf, and its function is to convert the measured current I_in into a voltage V_out1 ≈ -I_in x Rf, and to improve the loop phase margin through Cf to suppress high-frequency instability caused by high Rf.

[0071] The value range of Rf is 1 GΩ-100 GΩ, and the value range of Cf is 1 pF-10 pF.

[0072] In engineering, Cf can be selected according to the equivalent capacitance C_IN of the measured object and the input cable to make the system phase margin ≥ 45°; the experience starting point is Rf = 100 GΩ, Cf = 3.3 pF, and the step response is fine-tuned to the interval of 3.3-6.8 pF.

[0073] Stand column suspension connection: the inverting end high resistance node and the feedback device are connected by a stand column suspension connection for air suspension direct connection, specifically: a PTFE or POM insulating stand column (diameter 3-4 mm, height 10-15 mm) is used to support the feedback resistor / capacitor integrated piece; The inverting end pin is welded to the suspended end of the integrated piece through a gold-plated short lead of ≤ 8 mm; ceramic gaskets and a small amount of non-silicon-based glue are used for positioning; the shortest creepage distance from the node to the FR-4 board surface is controlled to be ≥ 8 mm.

[0074] This structure makes the high resistance node away from moisture and pollutants, significantly reducing the surface leakage channel.

[0075] 4.2 Guard ring buffer and guard ring conductor The guard ring conductor is a ring-shaped or closed strip-shaped copper band (width 2-4 mm) arranged around the inverting end high resistance node, and is ≥ 1.5 mm away from any high resistance metal piece.

[0076] The guard ring buffer is based on the principle of low output impedance and unit gain stability, with a bandwidth ≥ 100 kHz.

[0077] During normal measurement, the guard ring buffer drives the guard ring conductor to the same potential as the high impedance node, suppressing displacement current from the dielectric and parasitic capacitance; during the window period, the digital controller superimposes a 5 mV to 50 mV micro-excitation step to the guard ring through a programmable micro-stepping potential generator, to excite the response of the leakage channel for the estimation of the equivalent leakage conductance.

[0078] 4.3 Inner and outer double-layer metal shielding structure The shielding scheme adopts two layers: the inner layer shielding connects to the analog ground, mainly shielding the high impedance area of the operational amplifier and feedback network; the outer layer shielding connects to the chassis ground, forming an electromagnetic barrier to the outside.

[0079] The two layers are isolated by an insulating gasket, and are coupled to the chassis ground at a single point through a capacitor or a resistor-capacitor network to suppress circulating current and ground potential difference.

[0080] The shielding box gap adopts a stepped lap joint, with a lap width ≥ 5 mm and a screw spacing ≤ 40 mm at the four corners and the edge, to ensure shielding continuity and gap coupling suppression.

[0081] 4.4 Input and interface (supporting SMA / three coaxial) The input uses a shielded connector. Preferably, a three-coaxial interface (inner core = signal, inner shield = guard ring, outer shield = chassis ground) is used, which can minimize cable medium leakage.

[0082] When SMA compatibility is required, a special guard ring adapter is provided: the SMA outer conductor is driven to the same potential as the high impedance node by the guard ring buffer; at the same time, the independent outer shield is connected to the chassis ground, avoiding the contradiction of being driven and grounded in the same layer.

[0083] 4.5 RC low-pass filter and post-amplification / switching A first-order RC low-pass filter (example: R_LP=1 kΩ, C_LP=1 µF, corresponding to fc≈160Hz) is connected in series after the transimpedance output.

[0084] Then enter the second-stage zero-drift amplifier and analog switch array, which can select G2=1, 2, 5, 10 for the post-voltage gain.

[0085] When the ADC front-end voltage is close to 80% of the full scale, the digital controller automatically bypasses the second-stage amplification to expand the scale and avoid saturation distortion.

[0086] The analog switch uses an ultra-low leakage type (I_leak≤50 pA@25 °C).

[0087] 4.6 Analog-to-digital converter and reference The analog-to-digital converter adopts a 24-bit sigma-delta architecture, has a sampling rate of ≥2 kSPS, differential input, and an internal or external PGA that can be set to 1-8 times.

[0088] The reference voltage is 2.500 V, and the temperature drift is ≤10 ppm / °C; RC filtering is added between the reference and the ADC power supply to isolate digital noise.

[0089] The secondary amplification output is connected to the ADC differential input through a simple RC to ensure driving capability and stability; during sampling, a timer is triggered, and a programmable integration window and sliding average are used to suppress low-frequency 1 / f noise.

[0090] 4.7 Digital Controller and Programmable Micro-stepping Potential Generator The digital controller (MCU or SoC) has a main frequency of ≥120 MHz, with DMA and multi-channel timers; a built-in or external 16-bit DAC is used for the micro-stepping potential generator output.

[0091] The controller manages the "normal sampling window / non-sampling window (empty window)" in beats: only during the empty window, the micro-excitation step is output to the guard ring conductor; after the micro-excitation ends, the sampling of the measured input is restored.

[0092] The refresh period T can be configured to 1-60 s, and the default value in this embodiment is 5 s.

[0093] 4.8 Environment Perception Module The environment perception module has built-in temperature and humidity sensors with accuracies of not less than ±0.2 °C and ±2%RH, respectively, and a range covering -10-60 °C and 10-90%RH.

[0094] The module is close to the input interface and the guard ring area to improve the accuracy of the actual leakage channel environment perception, and at the same time, it maintains a distance of ≥15 mm from the high resistance node to avoid pollution and parasitism.

[0095] The sensor data is periodically sent to the digital controller for correction of the temperature and humidity-leakage admittance model and zero point / proportion factor.

[0096] Five, Working Principle and Mathematical Model (Including Formulas) (1) Transimpedance amplification relationship: under low-frequency approximation, the transimpedance output voltage V_out1 ≈ - I_in × R_f.

[0097] The parallel compensation capacitor C_f and the total node capacitance C_T (including the measured object and cable capacitance C_IN and parasitic) jointly determine the pole and zero positions, ensuring the loop phase margin.

[0098] (2) Shield micro-excitation and leakage estimation: Apply a micro-step AV_g (5 mV~50 mV) to the shield conductor during the window period, and the equivalent leakage admittance G_leak around the high-impedance node produces a current change ΔI = G_leak × ΔV_g.

[0099] The step response of the transimpedance output to this change is approximately ΔV_out1(t) = - R_f × ΔI × (1 - e^{-t / τ}), where τ ≈ R_f × (C_T +C_f).

[0100] When the step lasts t_s ≥ 3τ, it is considered to enter the steady state: ΔV_out1,ss ≈ - R_f × G_leak × ΔV_g.

[0101] After RC low-pass and two-stage gain, it reaches the ADC, and the digital controller first or second order difference of the sampling sequence to get the measurement value ΔV_meas, according to which to estimate G_leak = - ΔV_meas / (K × R_f × ΔV_g), Where K is the proportionality coefficient determined by RC low-pass, two-stage gain and ADC range (calibration section).

[0102] (3) Online compensation: Under normal sampling window, the measured current I_meas = I_in + I_leak, Where the leakage component I_leak ≈ G_leak × (V_sur - V_node). Under the shield potential, (V_sur - V_node) is small, but residual non-isopotential path still exists.

[0103] Using the estimated value G_leak, the controller calculates I_corr = I_meas - Î_leak, Get the output current I_corr after online compensation.

[0104] Six, mechanical and process implementation (column suspension and shielding assembly) (1) Base plate material: FR-4 is used for digital / control area, and low water absorption laminated board is preferred for analog front end; high-impedance node around the forbidden cover solder and character ink.

[0105] (2) Column suspension: PTFE column (diameter 3-4 mm) is installed in combination with ceramic gasket, height 10-15 mm, to ensure the shortest creepage distance ≥8 mm; feedback resistor / capacitor integrated component is fixed on the top of the column; short lead of ≤8 mm of the inverting end is welded to the suspension end of the integrated component.

[0106] (3) Guard ring conductor: Independent small plate or copper strip, suspended and fixed by an insulating bracket; the connection line from the guard ring buffer to the guard ring is ≤20 mm to avoid phase delay and ringing.

[0107] (4) Assembly of inner and outer shielding: First install the digital board and outer shielding, then install the analog board, pillars and protective rings, inner shielding cover, and finally connect the plugs and close the chassis; the outer chassis and inner shielding are coupled by capacitor or resistor-capacitor single-point coupling (typically R=100Ω, C=100 nF).

[0108] (5) Cleaning and drying: After welding, clean with deionized water and isopropanol, and dry with hot air at 60 °C for ≥30 min; silicone additives must not be left behind.

[0109] VII. Firmware Architecture and Algorithm Flow (Can be implemented directly) State machine process: S0 initialization → S1 normal sampling → S2 micro-excitation during the window period → S3 leakage estimation → S4 parameter refresh (including temperature and humidity model) → S5 range and integral window adaptation → S6 output and recording.

[0110] Sampling of the input being tested is paused during S2 and resumed immediately afterward. The refresh period T can be set from 1 to 60 seconds.

[0111] Pseudocode (core): loop: if time_to_measure: I_meas = measure_with_integration() # Programmable integral window + moving average if time_to_guard_pulse: pause_input_sampling() apply_guard_step(ΔVg=10mV, ts=2ms) ΔVmeas = capture_response_and_diff() resume_input_sampling() Ghat = -ΔVmeas / (K * Rf * ΔVg) if time_to_refresh(T=5s): (T_env, H_env) = read_env() Ghat = env_model_update(Ghat, T_env, H_env) (zero, scale) = update_zero_and_scale(T_env, H_env) I_corr = I_meas - I_leak_est(Ghat) auto_range_and_window(I_corr) # auto range and window the secondary amplifier output_record(I_corr, T_env, H_env, Rf, G2, window, Ghat) Eight, calibration and parameter acquisition (not less than 80 points of linearity calibration) (1) Linearity calibration: use standard picoampere current source or transimpedance feedback to cover each range of the range with ≥80 equidistant points; obtain (I_set, V_ADC), and use piecewise linear / polynomial fitting to generate linearization LUT and store in the controller.

[0112] (2) Zero point calibration: input short or internal short circuit, integrate 1 s, collect ≥5 min average value as zero point, store in NVRAM.

[0113] (3) Proportion factor: fit at several standard current points (such as ±1 nA, ±5 nA, ±10 nA) to obtain scale factor and temperature and humidity correction term.

[0114] (4) Leakage estimation proportion coefficient K: under the condition of input open circuit and normal operation of the guard ring, apply ΔVg step with known equivalent leakage device (such as 10 TΩ and 2 pF), measure ΔVmeas to obtain K; calibrate K_i for different Rf / Cf / G2 combinations and build a table.

[0115] (5) Temperature and humidity empirical model: record multiple sets of (T, H, G_leak) data, fit G_leak(T,H) = a0 + aT·(T-25) + aH·(H-40) + aTH·(T-25)(H-40), Store the coefficients and update them smoothly in operation.

[0116] Nine, test items and acceptance criteria (1) Power up and interface: Check if ±5 V, +3.3 V, +2.5 V references are within ±1%, ADC clock and trigger synchronization are normal; three coaxial / SMA interfaces are not loose.

[0117] (2) Guard ring stability: oscilloscope monitor guard ring and high resistance node AC difference <2 mV_rms; if out of limit, check buffer ringing and cable length.

[0118] (3) Phase margin verification: step response overshoot <15%, if out of limit, appropriately increase C_f (for example, from 3.3 pF to 4.7 pF).

[0119] (4) RC low pass verification: with 1 kHz square wave micro-current injection, out-of-band suppression >30 dB.

[0120] (5) Leakage estimation link: ΔVg=10 mV, t_s=2 ms, ΔVmeas is stable and repeatable, K value falls within the calibration interval ±10%.

[0121] (6) Linearity: after LUT correction, linearity within ±10 nA range is better than ±0.01%.

[0122] (7) Resolution and noise: Rf=100 GΩ, integration 1 s, noise ≤5 fA_rms.

[0123] (8) Temperature and humidity drift: 25→35 °C, 40→70%RH step, after compensation, zero point drift is ≤20 fA and ≤50 fA, respectively.

[0124] (9) Long-term operation: 168 h continuous operation, zero point jitter ≤30 fA, automatic bypass function and adaptive integration window work normally.

[0125] Ten, typical measured data (example, not limited to the scope of protection) Under the conditions of Rf=100 GΩ, C_f=4.7 pF, G2=5, integration time 1 s: — Noise: 4.6 fA_rms; — Linearity: better than ±0.01% within ±10 nA range; — Temperature step (25→35 °C): without compensation, zero point offset is about +120 fA, after compensation, it is about +18 fA; — Humidity step (40→70%RH): without compensation, zero point offset is about +320 fA, after compensation, it is about +45 fA; — Refresh period T=5 s, continuous operation for 168 h, G_leak fluctuation ±12%, I_corr zero point jitter ≤30 fA.

[0126] The above data demonstrate the high resolution, strong anti-interference and long-term stability of the present embodiment in pascal level measurement.

[0127] Eleven, safety, maintenance and use recommendations (1) ESD and transport protection: insert 100 kΩ protection resistor at input end and add empty end cap during transport / maintenance; remove the protection before measurement to avoid affecting the transimpedance accuracy.

[0128] (2) Cleaning period: it is recommended to perform internal cleaning and zero recalibration once every 1000 hours or every year; the period should be appropriately shortened in high humidity or dusty environments.

[0129] (3) Firmware upgrade: online update through Bootloader; after the upgrade is completed, automatically perform zero quick re-calibration and K value self-checking.

[0130] (4) Cable selection: prefer three coaxial; when using SMA, make sure to use a guard ring adapter to avoid introducing leakage current by directly grounding the outer conductor.

[0131] (5) Environment: avoid condensation; if necessary, increase the desiccant or micro-heating to stabilize the humidity.

[0132] The above only describes the preferred embodiments of the present application, and does not limit the patent scope of the present application, and any equivalent structural transformation made under the inventive concept of the present application, using the content of the present application specification and drawings, or directly / indirectly applied in other related technical fields are included in the protection scope of the present application.

Claims

1. A picoampere-level current measuring device with leakage estimation and digital compensation, characterized in that, Comprise: An electrically metered operational amplifier and its trans-impedance feedback network for converting the measured current into voltage; The feedback network contains a feedback resistor of 1GΩ~100GΩ and a parallel compensation capacitor; The inverting input of the operational amplifier and the feedback network are connected by a column suspension connection for air suspension direct connection, so that the high resistance node of the inverting terminal is not in direct contact with the FR-4 board surface; A guard ring conductor driven by a guard ring buffer is arranged around the high resistance node of the inverting terminal; The device adopts a double-layer metal shielding structure, wherein the inner layer shielding is connected to the analog ground, and the outer layer shielding is connected to the chassis ground; The input end is provided with a shielded connector and supports a three-coaxial interface; The current measuring device further comprises an environment sensing module, an analog-to-digital converter and a digital controller, and the digital controller injects a micro-excitation of 5mV~50mV in amplitude into the guard ring conductor in a non-sampling window and measures the trans-impedance output response to estimate the input equivalent leakage conductance and compensate the measured current online.

2. The apparatus of claim 1, wherein, The compensation capacitor is 1pF~10pF to ensure the phase margin of the trans-impedance amplifier, and an RC low-pass filter is configured at the output end to suppress high-frequency noise and out-of-band interference.

3. The apparatus of claim 1, wherein, The column suspension connection includes a rigid column and an insulating support; the inverting terminal pin is directly welded to the suspended end of the feedback resistor / capacitor integrated component through a short lead, and the integrated component is fixed by the insulating support, so that the shortest creepage path from the inverting node to the board surface is greater than 8mm.

4. The apparatus of claim 1, wherein, The micro-excitation is controlled by a programmable micro-step potential generator of the digital controller and added to the guard ring conductor through the guard ring buffer; the micro-excitation step duration is 0.5ms~5ms, and the sampling of the measured input is suspended during the duration; The digital controller estimates the equivalent leakage conductance according to the first or second order difference of the output response, and updates the compensation parameters with an adaptive refresh period T=1s~60s.

5. The apparatus of claim 1, wherein, An insulating gasket is arranged between the inner layer and the outer layer of the double-layer metal shielding structure; the outer layer shielding is directly connected to the chassis ground, and the inner layer shielding is coupled to the chassis ground at a single point through a capacitor or a resistor-capacitor network to suppress circulating current; the shielding box gap adopts a stepped lap joint to reduce the gap coupling.

6. The apparatus of claim 1, wherein, The input end can be configured with an SMA interface or a three-coaxial interface; when the SMA interface is selected, the SMA outer conductor is made equal to the high resistance node by the low output impedance drive of the guard ring buffer through a special guard ring adapter, and is connected to the chassis ground through an independent outer layer shielding to reduce cable leakage and maintain the integrity of the chassis shielding.

7. The apparatus of claim 1, wherein, A switchable two-stage zero drift amplifier and an analog switch array are connected in series after the trans-impedance output to realize the selection of two-stage gain G2=1, 2, 5, 10; when the output approaches saturation, the two-stage amplifier is automatically bypassed to expand the linear range.

8. The apparatus of claim 1, wherein, The environment sensing module includes temperature and humidity sensors; the digital controller performs temperature and humidity compensation on the zero point and the scale factor based on the temperature and humidity-leakage admittance empirical model.

9. The apparatus of claim 1, wherein, The analog-to-digital converter is 24-bit with a sampling rate of ≥2kSPS; combined with segmented multi-point (not less than 80 points) linearity calibration and programmable integration window, the low-frequency 1 / f noise is suppressed by sliding average or window integration.

10. A method for measuring a picoampere level current with leakage estimation and digital compensation, applied to the device of any one of claims 1-9, characterized in that, The method comprises: S1, setting the trans-impedance feedback parameters and the compensation capacitor value; S2, collect the trans-impedance output voltage and convert the current according to the calibration coefficient; S3, inject micro-excitation to the guard ring conductor in the non-sampling window and measure the output response, suspend the sampling of the measured input during the micro-excitation, and estimate the input equivalent leakage conductance; S4, compensate the current data on-line according to the leakage estimation and the temperature and humidity model; S5, adaptively select the secondary gain and the integration window according to the output amplitude, so as to balance the resolution and the dynamic range.

Images