A parameter self-calibration method, digital ultrasonic generator and wire bonding equipment

Abstract

The application relates to the technical field of ultrasonic transducer driving and control, in particular to a parameter self-calibration method, a digital ultrasonic generator and a wire bonding device. The method controls the digital ultrasonic generator to perform resonance capture from a first initial driving frequency to the resonance state of an ultrasonic transducer system, and obtains a first resonance capture track; a reference resonance frequency is determined according to the first resonance capture track, and second resonance capture is performed from the other side of the reference resonance frequency, and a second resonance capture track is obtained; according to the electrical response in the two tracks and the difference in the capture direction, the response of the transducer system irrelevant to the resonance capture direction is determined; the transducer system parameters are identified according to the transducer system response and a preset parameter model, and the digital ultrasonic generator is parameter self-calibrated according to the transducer system parameters. The application can complete parameter identification and self-calibration by using actual resonance capture data, reduce the influence of the capture direction and the frequency change process on the identification result, and reduce the dependence on independent and complete sweep frequency tests.

Description

Technical Field

[0001] This application relates to the field of ultrasonic transducer driving and control technology, specifically to a parameter self-calibration method, a digital ultrasonic generator, and a wire bonding device. Background Technology

[0002] In ultrasonic processing equipment such as wire bonding, a digital ultrasonic generator typically outputs an electrical drive signal with a preset frequency and amplitude to the ultrasonic transducer system. The ultrasonic transducer then converts the electrical energy into mechanical ultrasonic vibrations and transmits them to the actuator. To ensure the ultrasonic transducer system operates at a suitable resonant state, the digital ultrasonic generator can acquire the voltage and current signals of the ultrasonic transducer system. Based on the amplitude, phase, or impedance relationship between the voltage and current, the resonant state of the transducer system is determined, and the drive frequency is adjusted through closed-loop control methods such as phase-locked loops.

[0003] However, the actual frequency response of an ultrasonic transducer system is affected by factors such as the transducer's own parameters, the installation status of the actuator, mechanical load, and connection and detection links. For example, the resonant frequency, impedance characteristics, and phase characteristics of the transducer system may change after using different transducers, different connecting cables, or replacing the actuator. Therefore, to improve the compatibility between digital ultrasonic generators and current transducer systems, existing technologies have proposed frequency sweeping or frequency search based on voltage and current responses to determine the resonant frequency, impedance, or equivalent parameters of the transducer.

[0004] The above parameter identification methods typically require acquiring the electrical responses at multiple frequency points within a preset frequency range. If identification data is acquired through an independent frequency sweep process, a parameter testing process needs to be added in addition to the normal resonant tracking process. If, in order to shorten the calibration time, the electrical response generated during the closed-loop capture process from the initial driving frequency to the resonant state using a digital ultrasonic generator is directly utilized, the driving frequency will continuously change during the resonant capture process, and the electrical responses obtained under different capture directions may be simultaneously affected by the frequency characteristics of the transducer system, as well as factors such as the capture direction and frequency change process.

[0005] In particular, for two resonant capture processes converging from different lateral resonant states at resonant frequencies, even if the two captures pass through the same or adjacent driving frequencies, their corresponding electrical responses may differ. If the dynamic electrical response obtained during a single resonant capture process is directly used as the basis for identifying transducer system parameters, response differences related to the capture direction or frequency change process may be introduced into the identification results of the transducer system's frequency characteristics. If a relatively stable frequency response is still obtained through additional complete frequency sweep tests, it is difficult to fully utilize the electrical response data already generated during the normal resonant capture process of the digital ultrasonic generator.

[0006] Therefore, how to utilize the electrical response generated during the resonant capture process of a digital ultrasonic generator to reduce the impact of the resonant capture direction and its frequency variation on the identification of transducer system parameters, while simultaneously achieving parameter self-calibration of the digital ultrasonic generator, is a technical problem that needs to be solved. Summary of the Invention

[0007] This application provides a parameter self-calibration method, a digital ultrasonic generator, and a wire bonding device to solve the problem that existing digital ultrasonic generators are unable to fully utilize the electrical response data generated during the resonant capture process while reducing the impact of the resonant capture direction and its frequency variation process on the transducer system parameter identification results.

[0008] To address the aforementioned technical problems, this application provides a parameter self-calibration method in a first aspect, applied to a digital ultrasound generator, comprising: controlling the digital ultrasound generator to resonate from a first initial driving frequency to the resonant state of an ultrasound transducer system, and acquiring a first resonant capture trajectory; determining a reference resonant frequency based on the first resonant capture trajectory, and determining a second initial driving frequency such that the first initial driving frequency and the second initial driving frequency are located on opposite sides of the reference resonant frequency; resonating from the second initial driving frequency to the resonant state to acquire a second resonant capture trajectory with a different capture direction than the first resonant capture trajectory; determining a transducer system response independent of the resonant capture direction based on the electrical responses in the first and second resonant capture trajectories and the difference in capture direction between the two resonant capture trajectories; identifying transducer system parameters based on the transducer system response and a preset transducer system parameter model, and performing parameter self-calibration of the digital ultrasound generator based on the transducer system parameters.

[0009] Furthermore, controlling the digital ultrasound generator to perform resonance capture includes: acquiring the voltage response and current response of the ultrasound transducer system, determining the phase error based on the phase difference between the voltage response and the current response, and adjusting the driving frequency of the digital ultrasound generator in a closed loop based on the phase error so that the driving frequency converges to the resonance state of the ultrasound transducer system; both the first resonance capture trajectory and the second resonance capture trajectory are formed by the actual closed-loop adjustment process of the driving frequency.

[0010] Furthermore, the first resonant capture trajectory and the second resonant capture trajectory each include multiple trajectory points, each trajectory point including a driving frequency, a frequency variation characteristic representing the change of the driving frequency over time, and a corresponding electrical response; the determination of the transducer system response independent of the resonant capture direction includes: establishing a trajectory data association between the two resonant capture trajectories based on the driving frequency, frequency variation characteristic, and electrical response corresponding to each trajectory point in the first and second resonant capture trajectories, and determining a response component with consistency under different capture directions based on the trajectory data association, as the transducer system response.

[0011] Furthermore, the frequency change characteristics include the frequency change rate; a preset identification frequency range is determined near the reference resonant frequency; during the first resonance capture process and the second resonance capture process, the driving frequency is controlled to cross the preset identification frequency range along opposite frequency change directions, so that both the first resonance capture trajectory and the second resonance capture trajectory cover the preset identification frequency range; after the first resonance capture process and the second resonance capture process have completed crossing the preset identification frequency range, the driving frequency is controlled to return to the reference resonant frequency to enter the resonance state.

[0012] Furthermore, multiple common driving frequencies are determined within the preset identification frequency range; for each common driving frequency, corresponding trajectory data are determined from the first resonant capture trajectory and the second resonant capture trajectory, respectively; when there is a trajectory point in the resonant capture trajectory that is the same as the common driving frequency, the trajectory data corresponding to the common driving frequency is determined according to the frequency change rate and electrical response corresponding to the trajectory point; when there is no trajectory point in the resonant capture trajectory that is the same as the common driving frequency, interpolation is performed according to the driving frequency, frequency change rate and electrical response corresponding to the adjacent trajectory points located on both sides of the common driving frequency to obtain the frequency change rate and electrical response under the common driving frequency.

[0013] Furthermore, the electrical responses originating from the first and second resonant capture trajectories at each of the aforementioned common driving frequencies are respectively represented as a combination of a direction-independent response component corresponding to the current common driving frequency and independent of the resonant capture direction, and a direction-dependent response component whose characteristics change with the corresponding frequency. The direction-independent response component is used as the transducer system response. Based on the frequency variation characteristics and electrical responses originating from the first and second resonant capture trajectories at the same common driving frequency, the electrical response is decomposed according to the combination relationship to determine the direction-independent response component and the direction-dependent response component.

[0014] Furthermore, the direction-dependent response component has a preset linear relationship with the frequency change rate; for the first trajectory data and the second trajectory data respectively derived from the first resonant capture trajectory and the second resonant capture trajectory under the same common driving frequency, the direction-independent response component and the dynamic response parameter characterizing the degree of change of the direction-dependent response component with the frequency change rate are solved according to the electrical response and frequency change rate corresponding to the first trajectory data and the second trajectory data respectively, as well as the preset linear relationship.

[0015] Furthermore, during the formation of at least one of the first resonant capture trajectory and the second resonant capture trajectory, a phase error is determined based on the phase difference between the voltage response and the current response; when the phase error reaches a preset phase error value or enters a preset phase error range, the corresponding trajectory point is determined as an identification anchor point; for the identification anchor point, the current driving frequency is maintained for a preset number of driving cycles, and the electrical response corresponding to the identification anchor point is updated based on the voltage response and current response obtained during the maintenance of the driving frequency, and then the resonant capture continues.

[0016] Furthermore, the voltage response and the current response are obtained through a sampling process synchronized with the current driving frequency; the voltage signal and the current signal of the ultrasonic transducer system are sampled respectively during at least one complete driving cycle, and the obtained voltage sample value and current sample value are processed in the frequency domain to determine the voltage complex response and the current complex response at the corresponding driving frequency; the electrical response at the corresponding driving frequency is determined based on the voltage complex response and the current complex response, and the electrical response includes at least one of complex impedance and complex admittance.

[0017] Furthermore, the transducer system parameter model includes a dynamic branch characterizing the electromechanical dynamic characteristics of the ultrasonic transducer system and a parasitic branch connected in parallel with the dynamic branch; the transducer system parameters include the equivalent dynamic resistance, equivalent dynamic inductance, and equivalent dynamic capacitance corresponding to the dynamic branch, and the equivalent parallel parasitic capacitance corresponding to the parasitic branch; the parameter self-calibration method further includes: during the formation of at least one of the first resonant capture trajectory and the second resonant capture trajectory, acquiring the voltage response and current response of the ultrasonic transducer system, and identifying detection link parameters characterizing the response difference between the voltage detection link and the current detection link based on the voltage response and the current response; the detection link parameters include at least one of the relative gain parameter and the relative time delay parameter between the voltage detection link and the current detection link.

[0018] Furthermore, the parameter self-calibration includes at least one of the following: determining the parasitic current component based on the identified equivalent parallel parasitic capacitance, and compensating for the current response used for resonant capture based on the parasitic current component; correcting the phase relationship between the voltage response and the current response based on the relative time delay parameter when the relative time delay parameter is identified; determining the estimated resonant frequency of the ultrasonic transducer system based on the equivalent dynamic inductance and the equivalent dynamic capacitance, and configuring the initial drive frequency for the subsequent resonant capture process based on the estimated resonant frequency; configuring resonant tracking control parameters for closed-loop adjustment of the drive frequency based on at least one of the equivalent dynamic resistance, the equivalent dynamic inductance, and the equivalent dynamic capacitance; and configuring the ultrasonic output control parameters of the digital ultrasonic generator based on the impedance characteristics characterized by the transducer system parameters.

[0019] This application provides a digital ultrasound generator with parameter self-calibration function in a second aspect, including an ultrasound drive module, an electrical response acquisition module, and a digital control module; the ultrasound drive module is used to output an ultrasound drive signal to an ultrasound transducer system; the electrical response acquisition module is used to acquire the voltage response and current response of the ultrasound transducer system; the digital control module is used to determine the electrical response of the ultrasound transducer system based on the voltage response and the current response, and is used to: control the ultrasound drive module to perform resonance capture from a first initial drive frequency to the resonant state of the ultrasound transducer system, and acquire a first resonance capture trajectory; determine a reference resonant frequency based on the first resonance capture trajectory, and determine a second initial... The initial driving frequency is set such that the first initial driving frequency and the second initial driving frequency are located on opposite sides of the reference resonant frequency. The ultrasonic driving module is controlled to resonate from the second initial driving frequency to the resonant state to obtain a second resonant capture trajectory with a different capture direction than the first resonant capture trajectory. Based on the electrical responses in the first and second resonant capture trajectories and the difference in capture direction between the two resonant capture trajectories, a transducer system response independent of the resonant capture direction is determined. Based on the transducer system response and a preset transducer system parameter model, the transducer system parameters are identified, and the digital ultrasonic generator is self-calibrated based on the transducer system parameters.

[0020] Furthermore, the ultrasonic drive module includes a direct digital frequency synthesis unit, a digital-to-analog converter, an amplitude adjustment unit, and a power amplification unit arranged sequentially. The direct digital frequency synthesis unit is used to generate a frequency-adjustable digital drive signal. The digital drive signal is processed by the digital-to-analog converter, the amplitude adjustment unit, and the power amplification unit and then output to the ultrasonic transducer system. The electrical response acquisition module includes a voltage detection unit, a current detection unit, and an analog-to-digital converter. The voltage detection unit and the current detection unit are used to acquire the voltage signal and the current signal of the ultrasonic transducer system, respectively. The digital control module is used to control the analog-to-digital converter to synchronously sample the voltage signal and the current signal according to the current drive frequency, and to determine the voltage response and the current response based on the synchronous sampling results.

[0021] This application provides a wire bonding apparatus in a third aspect, including a digital ultrasonic generator and an ultrasonic transducer system connected to the digital ultrasonic generator; the ultrasonic transducer system includes an ultrasonic transducer and a bonding tool mechanically connected to the ultrasonic transducer; the digital ultrasonic generator is used to perform resonance capture from a first initial driving frequency to the resonant state of the ultrasonic transducer system to obtain a first resonance capture trajectory, determine a reference resonant frequency based on the first resonance capture trajectory, and determine a second initial driving frequency, such that the first initial driving frequency and the second initial driving frequency are respectively located on both sides of the reference resonant frequency; and to perform resonance capture from the second initial driving frequency to the resonant state to obtain a resonance capture trajectory. The resonant capture trajectory is a second resonant capture trajectory with a different capture direction; the digital ultrasonic generator is also used to acquire the voltage and current responses of the ultrasonic transducer system during the first and second resonant capture processes, and determine the corresponding electrical responses based on the voltage and current responses; based on the electrical responses in the first and second resonant capture trajectories and the difference in capture direction between the two resonant capture trajectories, determine the transducer system response independent of the resonant capture direction, identify the transducer system parameters based on the transducer system response and a preset transducer system parameter model, and configure the ultrasonic control parameters of the digital ultrasonic generator during wire bonding based on the transducer system parameters.

[0022] The above technical solution has the following advantages compared to the existing technology: This application obtains a first resonant capture trajectory by performing actual closed-loop resonant capture from a first initial driving frequency to the resonant state of the ultrasonic transducer system. A reference resonant frequency is determined based on this trajectory, and a second actual closed-loop resonant capture is performed from the other side of the reference resonant frequency to obtain a second resonant capture trajectory with different capture directions. Furthermore, the transducer system response, independent of the resonant capture direction, is determined based on the electrical response data generated during the two resonant capture trajectories and the difference in capture direction. Therefore, while using the electrical response data generated during resonant capture for transducer system parameter identification, the influence of resonant capture direction and frequency changes on the parameter identification results is reduced. The digital ultrasonic generator can also be self-calibrated based on the identified transducer system parameters.

[0023] The above technical solution also has the following advantages: Both the first and second resonant capture trajectories can be formed by the actual closed-loop control process of the digital ultrasonic generator. This allows the resonant capture process to simultaneously provide the electrical response data required for transducer system parameter identification, helping to reduce reliance on independent and complete frequency sweep testing processes. Based on the obtained transducer system parameters, resonant tracking control parameters, ultrasonic output control parameters, parasitic current compensation parameters, or the initial drive frequency for subsequent resonant captures can be further configured to adapt the digital ultrasonic generator to the frequency characteristics of the current ultrasonic transducer system. Attached Figure Description

[0024] Figure 1 A flowchart illustrating a parameter self-calibration method for a digital ultrasound generator provided in this application embodiment; Figure 2 A schematic diagram of the structure of a digital ultrasonic generator, an ultrasonic transducer system, and a wire bonding device provided in an embodiment of this application; Figure 3 A schematic diagram showing the first and second resonant capture trajectories provided in this application embodiment traversing a preset identification frequency range along opposite frequency change directions and returning to the resonant lock state; Figure 4 A schematic diagram illustrating the principle of electrical response decomposition based on trajectory data corresponding to different resonance capture directions, provided in an embodiment of this application; Figure 5 This is an equivalent circuit diagram of a parameter model of an ultrasonic transducer system provided in an embodiment of this application.

[0025] Explanation of reference numerals in the attached figures: 100. Digital ultrasonic generator; 110. Ultrasonic drive module; 111. Direct digital frequency synthesis unit; 112. Digital-to-analog conversion unit; 113. Amplitude adjustment unit; 114. Power amplification unit; 120. Electrical response acquisition module; 121. Voltage detection unit; 122. Current detection unit; 123. Analog-to-digital conversion unit; 130. Digital control module; 200. Ultrasonic transducer system; 210. Ultrasonic transducer; 220. Bonding tool; 300. Bonding object; 400. Wire bonding equipment. Detailed Implementation

[0026] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the following embodiments are used to illustrate this application, and not to limit the scope of protection of this application. Unless otherwise specified, the following embodiments and their technical features can be combined with each other. Conventional circuit connections, signal conditioning, data storage, and program execution methods that can be directly determined by those skilled in the art based on the description in this application will not be described in detail.

[0027] The digital ultrasonic generator described in this application is primarily used to drive an ultrasonic transducer system to generate mechanical ultrasonic vibrations. In wire bonding applications, the ultrasonic transducer system may include an ultrasonic transducer and a bonding tool mechanically connected to the ultrasonic transducer. The bonding tool may be, for example, a capillary tube, a chopping tool, or other tool capable of transmitting ultrasonic vibrations to the bonding site. The digital ultrasonic generator can also be applied to other devices that require resonant driving and parameter configuration of the ultrasonic transducer system. This application is not limited to specific application scenarios, execution tool types, or processing object types.

[0028] To facilitate understanding of the technical solution of this application, the following terms are explained first. "Resonance state" refers to the operating state of an ultrasonic transducer system where the driving frequency meets a preset resonance determination condition. The resonance determination condition can be determined based on at least one of the following: the phase relationship between the voltage and current responses, impedance or admittance characteristics, and the lock-in state of the resonance tracking controller. In one embodiment, the phase error is determined based on the phase difference between the voltage and current responses. When the phase error enters a preset lock-in range and persists for a preset number of driving cycles, the ultrasonic transducer system is determined to have entered a resonance state. The resonance state does not require the phase difference between the voltage and current responses to be strictly zero; a non-zero target phase relationship can be set depending on the detection link, parasitic parameters, and control objectives.

[0029] Let the actual phase difference between the voltage response and the current response be... The preset target phase difference is Then the phase error can be expressed as:

[0030] in, For phase error, The phase difference is determined by the voltage response and the current response. The preset target phase difference. When Less than or equal to the preset phase error threshold, or When the current driving frequency enters the preset phase error range, it can be considered that the current driving frequency meets one of the preset resonance determination conditions.

[0031] "Resonance capture" refers to the control process by which a digital ultrasonic generator, starting from an initial drive frequency, adjusts the drive frequency in a closed loop based on feedback values ​​characterizing the resonant state of the ultrasonic transducer system until the final resonance locking condition is met. For the parameter self-calibration mode of this application, a complete resonance capture process can sequentially include stages such as initial approach to the resonant state, acquisition of identification trajectory data, and re-entry into the final resonance locking state. Therefore, during a single resonance capture process, the drive frequency is allowed to traverse a limited range of identification frequencies near the reference resonant frequency after approaching it, before returning and locking back to the resonant state; this does not affect the fact that the complete process is called a resonance capture process.

[0032] A "resonance capture trajectory" refers to a set of multiple trajectory points recorded in chronological order during a single resonance capture process. These trajectory points can at least include the driving frequency and the corresponding electrical response, and preferably also include the sampling time, frequency variation characteristics, voltage response, current response, and phase error. A trajectory point can be represented as follows:

[0033] in, The time corresponding to the trajectory point, For driving frequency, The rate of change of frequency. and These are the voltage complex response and the current complex response, respectively. The electrical response is determined based on the complex voltage response and the complex current response. For phase error, This represents the trajectory point number. In other implementations, the trajectory point may only store the portion of data required for subsequent calculations, instead of storing all of the aforementioned data.

[0034] The "capture direction" characterizes the overall direction of change of the driving frequency relative to the reference resonant frequency during resonant capture. When the first initial driving frequency is located at a lower frequency side of the reference resonant frequency, the first resonant capture can generally approach the resonant state along the frequency-increasing direction; when the second initial driving frequency is located at a higher frequency side of the reference resonant frequency, the second resonant capture can generally approach the resonant state along the frequency-decreasing direction. The capture direction does not require the entire capture trajectory to be strictly monotonic at all times; local callbacks, amplitude limiting, and small-range adjustments during the final locking process are all possible.

[0035] "Frequency variation characteristics" are used to characterize the state of the driving frequency changing over time. These characteristics can include the direction of frequency change, the rate of frequency change, the frequency increment between adjacent trajectory points, or other quantities that can characterize the process of driving frequency change. The following uses the rate of frequency change as an example. The main implementation method will be described below. The frequency change rate can be obtained directly from the drive frequency command inside the controller, or it can be estimated differentially based on the frequency and time of adjacent trajectory points. For example, center difference can be used:

[0036] For the beginning and end points of the trajectory, one-sided differencing or local fitting can be used to obtain the frequency change rate. If necessary, low-pass filtering can be applied to the frequency sequence or frequency change rate to reduce the impact of quantization noise on subsequent response decomposition.

[0037] "Electrical response" is used to characterize the complex relationship between the voltage and current responses of an ultrasonic transducer system at a corresponding driving frequency. The following primarily uses complex admittance. Let's take an example to illustrate:

[0038] in, It is a voltage complex response. This is a complex current response. In other implementations, complex impedance can also be used:

[0039] Alternatively, other complex parameters that characterize the frequency response of the transducer system can be constructed from the complex voltage and current responses. The following pertains to the calculation of complex admittance, which can be transformed accordingly when using complex impedance as the electrical response.

[0040] "Direction-independent response components" refer to electrical response components that exhibit consistency relative to different resonant capture directions at the same common driving frequency. "Direction-dependent response components" refer to response components in the electrical response that vary with the capture direction and / or frequency. Direction-dependent response components can comprehensively reflect resonant capture dynamics, transducer nonlinearity, mechanical connection status, thermal state, and other influences related to the capture process; this application does not require them to be solely attributed to a single physical mechanism. Accordingly, the direction-independent response components described in this application are not limited to a strictly steady-state response or an absolutely true response devoid of any nonlinear effects, but rather refer to directionally consistent response components that can be extracted from data across different capture directions.

[0041] The "preset identification frequency range" is a frequency range determined based on the reference resonant frequency, used to acquire transducer system identification data under different resonant capture directions. It can be expressed as:

[0042] Preferred satisfaction:

[0043] in, As a reference resonant frequency, and These are the low-frequency and high-frequency boundaries of a preset identification frequency range. This range does not require... Being centrally symmetric, thus satisfying The interval width can be determined based on the estimated resonant bandwidth of the transducer system, the phase change range near the resonance, historical parameters, preset frequency deviation, or a combination of the above information.

[0044] The "common driving frequency" is a driving frequency located within a preset identification frequency range jointly covered by the first and second resonant acquisition trajectories, used to establish the correlation of trajectory data in different acquisition directions. Multiple common driving frequencies can be represented as... The common driving frequency can be determined according to a preset frequency interval, or it can be adaptively determined based on the actual trajectory points in a resonant capture trajectory.

[0045] Figure 2 This is a schematic diagram illustrating the structure of a digital ultrasonic generator, an ultrasonic transducer system, and a wire bonding device, provided as an embodiment of this application. See also... Figure 2 The wire bonding apparatus 400 may include a digital ultrasonic generator 100 and an ultrasonic transducer system 200 connected to the digital ultrasonic generator 100. The ultrasonic transducer system 200 may include an ultrasonic transducer 210 and a bonding tool 220 mechanically connected to the ultrasonic transducer 210, the bonding tool 220 being used to apply ultrasonic vibrations to the bonding object 300.

[0046] The digital ultrasonic generator 100 includes an ultrasonic drive module 110, an electrical response acquisition module 120, and a digital control module 130. The ultrasonic drive module 110 may include a direct digital frequency synthesis unit 111, a digital-to-analog converter 112, an amplitude adjustment unit 113, and a power amplifier unit 114 connected in sequence. The direct digital frequency synthesis unit 111 generates a frequency-adjustable digital drive signal based on the frequency control word or equivalent frequency control quantity provided by the digital control module 130. The digital drive signal is converted into an analog signal by the digital-to-analog converter 112, the drive amplitude is adjusted by the amplitude adjustment unit 113, and then amplified by the power amplifier unit 114 before being output to the ultrasonic transducer system 200.

[0047] The electrical response acquisition module 120 may include a voltage detection unit 121, a current detection unit 122, and an analog-to-digital converter 123. The voltage detection unit 121 acquires the voltage detection signal at the drive end of the ultrasonic transducer system 200, and the current detection unit 122 acquires the current detection signal flowing into the ultrasonic transducer system 200. The voltage and current detection signals are converted into digital sample values ​​by the analog-to-digital converter 123 and sent to the digital control module 130.

[0048] The digital control module 130 can be implemented using an FPGA, DSP, microcontroller, system-on-a-chip, or a combination of the above devices. In one embodiment, the digital control module 130 performs phase detection, resonance tracking, trajectory data recording, common drive frequency data construction, directional response decomposition, transducer system parameter identification, and parameter configuration. It should be noted that the above functions can be executed by the same processor or by multiple processors working together; it is not required that each function be set as an independent hardware module.

[0049] Figure 1 This is a schematic flowchart illustrating a parameter self-calibration method for a digital ultrasound generator provided in an embodiment of this application. See also... Figure 1 The parameter self-calibration method in this embodiment may include steps S101 to S106. The trajectory formation, trajectory association, response decomposition, parameter identification, and parameter self-calibration process in steps S101 to S106 will be further explained below.

[0050] Step S101: Obtain the first resonant capture trajectory. Specifically, the digital control module 130 controls the digital ultrasound generator 100 to perform the first resonant capture from the first initial driving frequency to the resonant state of the ultrasound transducer system 200. In one example, the first initial driving frequency is located on the low-frequency side of the target resonant mode, so the initial overall frequency change direction of the first resonant capture is the frequency increase direction. In another example, the first initial driving frequency can also be located on the high-frequency side, in which case the directional relationship described later is mirrored.

[0051] The digital control module 130 periodically acquires the voltage and current responses, calculates the phase difference between them, and obtains the phase error. Then, the driving frequency is adjusted in a closed-loop manner based on the phase error. Resonant tracking control can employ proportional-integral (PI) control, PI filtering control, or other closed-loop control algorithms. For example, discrete PI control can be used:

[0052] And according to:

[0053] Update the drive frequency for the next control cycle. Among these, Indicates the sequence number of the current control cycle. This indicates the control cycle index in integral operations. Indicates the first Each control cycle determines the frequency adjustment amount based on the phase error. and They represent the first The drive frequency of the first control cycle and the next control cycle. and These are the proportional and integral parameters for resonant tracking control, respectively. To control the cycle, This indicates that the amplitude is limited according to the allowed driving frequency range. The above proportional-integral relationship is only an example, and this application is not limited to any specific resonant tracking algorithm.

[0054] During the first resonance capture process, the digital control module 130 continuously records the driving frequency, voltage response, current response, frequency change characteristics, and corresponding electrical response, thereby gradually forming the first resonance capture trajectory.

[0055] Step S102: Determine the reference resonant frequency and the preset identification frequency range during the first resonant capture process.

[0056] In S102, the determination operation does not require waiting for the entire first resonance capture trajectory to be formed before execution. In a preferred embodiment, when the first resonance capture has formed a trajectory and first meets the preset preliminary resonance criterion, the reference resonant frequency used in this self-calibration cycle can be determined based on the obtained first trajectory data. The first resonant capture trajectory will then continue to form.

[0057] For example, when the phase error first enters the preset initial resonance range, the current driving frequency can be used as the reference resonance frequency; alternatively, the interpolated driving frequency can be selected near the zero-crossing of the phase error as the reference resonance frequency; or the reference resonance frequency can be determined by combining the phase error, admittance amplitude, and historical resonance frequency.

[0058] After obtaining the reference resonant frequency Then, determine the preset identification frequency range in the vicinity. For example, it can be based on:

[0059]

[0060] Determine the specific identification frequency range, where, , And the two can be equal or unequal. and It can be determined based on the estimated bandwidth of the target resonant mode, the phase slope near the resonance, historical identification parameters, or a preset maximum offset.

[0061] After determining the preset identification frequency range, it is controlled to traverse the preset identification frequency range in the first acquisition direction.

[0062] In order to ensure that the first resonant capture trajectory and the subsequent second resonant capture trajectory have a common frequency coverage range within the same identification frequency range, in the parameter self-calibration mode, the digital control module 130 does not immediately end the first resonant capture after the driving frequency enters the preset identification frequency range, but instead enters the identification trajectory acquisition state, so that the driving frequency continues to cross the preset identification frequency range in a limited manner along the current capture direction.

[0063] In one implementation, a bounded identification direction bias is superimposed on the resonant tracking control quantity:

[0064] in, This is the closed-loop resonant tracking control quantity obtained based on the phase error. This is a bounded identification bias that is enabled only when the identification trajectory is acquired. Indicates the direction of identification; the first capture can be taken when entering from the low-frequency side. The second capture can be obtained when it enters from the high-frequency side. The amplitude of the identification bias is limited to a preset range and is revoked after the driving frequency reaches the boundary of the corresponding identification interval or meets the preset identification termination condition. After the identification bias is revoked, the digital control module 130 resumes normal resonance locking control, causing the driving frequency to return to the resonance state. In a preferred embodiment, the magnitude of the identification bias can be adaptively determined based on the current closed-loop resonance tracking control quantity, so that the net frequency update in the identification trajectory acquisition state remains consistent with the current identification direction, while limiting the frequency change rate within a preset safety range.

[0065] In another implementation, instead of directly superimposing the frequency offset, the phase control target can be temporarily changed during trajectory acquisition, so that the target phase of the phase closure loop is changed from... Adjust the auxiliary target phase to one side of the current acquisition direction, so that the driving frequency can continue to traverse the identification frequency range while maintaining closed-loop control; restore it after completing the acquisition of identification trajectory data. This brings the driving frequency back to the final resonant locked state.

[0066] Therefore, the "crossing the preset identification frequency range" described in this application does not require the digital ultrasonic generator to perform an independent, complete frequency sweep test according to predetermined discrete frequency points. Instead, during a single closed-loop resonant capture process, the actual driving frequency trajectory is made to cover the preset identification frequency range through limited identification trajectory acquisition control. The frequency, time interval, and frequency change rate of the actual trajectory points can be naturally formed by the closed-loop control process and recorded as part of the first resonant capture trajectory.

[0067] Figure 3 This is a schematic diagram illustrating how the first and second resonant capture trajectories, provided in an embodiment of this application, traverse a preset identification frequency range along opposite frequency change directions and then return to the resonant lock state. See also... Figure 3 In the example where the first initial driving frequency is located on the low-frequency side of the reference resonant frequency, the first resonant capture trajectory enters and crosses the high-frequency boundary of the preset identification frequency range along the frequency increase direction. After obtaining the identification trajectory data corresponding to the first capture direction, the driving frequency is pulled back from the high-frequency side and finally locked near the reference resonant frequency.

[0068] Step S103: Determine the second initial driving frequency and acquire the second resonant capture trajectory.

[0069] After the first resonance capture is completed, the reference resonant frequency is determined or confirmed based on the first resonance capture trajectory. And determine the second initial drive frequency. This ensures that the first initial driving frequency and the second initial driving frequency are located on opposite sides of the reference resonant frequency. When the first initial driving frequency... When the second initial driving frequency is located on the low-frequency side of the reference resonant frequency, it is preferably located on the high-frequency side of the reference resonant frequency, and can be located on the high-frequency boundary of the preset identification frequency range. In addition, so that the second resonance captures the entry of the high-frequency side into the identification frequency range.

[0070] The first and second initial drive frequencies do not need to be symmetrical about the reference resonant frequency. For example, the second initial drive frequency can be selected based on the digital ultrasound generator's allowed start-up range, historical resonant frequencies, last calibration parameters, or a preset safe frequency deviation.

[0071] The digital control module 130 starts from a second initial driving frequency and adjusts the driving frequency in a closed loop based on the phase error between the voltage response and the current response. In one embodiment, during the second resonant acquisition process, the driving frequency enters the preset identification frequency range from the high-frequency side along the frequency reduction direction, and sequentially passes through the common driving frequency and the reference resonant frequency within the preset frequency range, until it crosses the low-frequency boundary of the preset frequency range. After obtaining the identification trajectory data corresponding to the second acquisition direction, the digital control module 130 cancels the identification trajectory acquisition control, causing the driving frequency to return from the low-frequency side to the reference resonant frequency and enter the final resonant locking state, thereby forming a continuous second resonant acquisition trajectory.

[0072] Therefore, within the preset identification frequency range, the first resonant capture trajectory and the second resonant capture trajectory have opposite overall frequency change directions. Preferably, at the trajectory points subsequently selected as paired identification data, the corresponding frequency change rates have opposite signs, that is:

[0073] in, For common drive frequency, and The first and second resonant capture trajectories represent the frequency change rates at the common driving frequency, respectively.

[0074] The following section explains the synchronous sampling of voltage and current and the calculation of electrical response.

[0075] To obtain an electrical response suitable for phase detection and parameter identification, in one embodiment, the sampling process of the analog-to-digital conversion unit 123 is synchronized with the drive signal generated by the direct digital frequency synthesis unit 111. For each trajectory point to be recorded, a voltage sample value is acquired within at least one complete drive cycle. and current sampling value .

[0076] For example, when each drive cycle acquires With equidistant sampling points, the fundamental voltage complex response and current complex response can be determined by discrete Fourier transform or synchronous demodulation:

[0077] in, The imaginary unit, This indicates the number of sampling points involved in the calculation within one driving cycle. Indicates the sampling point number. and They represent the first Each voltage sample value and current sample value. The above coefficients. The window function can be adjusted according to the actual sampling and window function form. When only the ratio of the voltage complex response to the current complex response is used subsequently, it is sufficient to use the same amplitude calibration method for both voltage and current.

[0078] In one example, 32 synchronous sampling points can be acquired within one drive cycle; in other implementations, other numbers of sampling points can be used, or coherent accumulation can be performed across multiple complete drive cycles to improve the signal-to-noise ratio. This application is not limited to the number of sampling points per cycle.

[0079] Calculate the complex admittance based on the complex voltage response and the complex current response:

[0080] And according to:

[0081] The phase difference between the voltage response and the current response is obtained, where, This represents the phase angle when taken as a complex number. When using complex impedance as the electrical response, the corresponding impedance phase relationship can be used accordingly.

[0082] The following section explains how to obtain the rate of frequency change and how to correlate trajectory data.

[0083] Both the first and second resonant capture trajectories can include multiple non-equally spaced trajectory points. Since the two captures are actually generated by the closed-loop control process, it is not required that the two trajectories be sampled at exactly the same time point or at exactly the same frequency point. Therefore, before response decomposition, multiple common driving frequencies can be determined within a preset identification frequency range, and trajectory data corresponding to each common driving frequency can be constructed from the two trajectories respectively.

[0084] For a certain common driving frequency If there exists a driving frequency equal to in the first resonant capture trajectory If the trajectory point is not specified, the frequency change rate and electrical response corresponding to that trajectory point are directly read; if there is no driving frequency exactly equal to... The trajectory points can then be determined based on their location. Interpolation is performed on adjacent trajectory points on both sides.

[0085] When the same resonant capture trajectory passes through the same common driving frequency multiple times due to the callback process after crossing, trajectory data can be selected based on the frequency change direction corresponding to the trajectory point. Specifically, for the first capture direction, trajectory points that are in the first direction identification trajectory acquisition stage and whose frequency change direction is consistent with the first capture direction are preferentially selected; for the second capture direction, trajectory points that are in the second direction identification trajectory acquisition stage and whose frequency change direction is consistent with the second capture direction are preferentially selected. Alternatively, multiple trajectory points that meet the preset frequency change direction conditions can be weighted.

[0086] Let the driving frequencies of two adjacent trajectory points on the same trajectory be respectively and And satisfy Define the interpolation coefficients:

[0087] The rate of change of frequency can be obtained by:

[0088] Complex admittance can be obtained through:

[0089] Obtained. Due to For complex numbers, the above complex linear interpolation is equivalent to performing linear interpolation on the real and imaginary parts separately. When using amplitude and phase to represent the electrical response, the phase can be made continuous first, and then the amplitude and phase can be interpolated separately.

[0090] The second resonance capture trajectory was obtained in the same way. and Therefore, for each common driving frequency, a pair of trajectory data can be constructed:

[0091] Data corresponding to multiple common driving frequencies together constitute the trajectory data association between the two resonant capture trajectories.

[0092] Step S104: Perform directional response decomposition based on bidirectional trajectory data.

[0093] Figure 4 This is a schematic diagram illustrating the principle of electrical response decomposition based on trajectory data corresponding to different resonance capture directions, as provided in an embodiment of this application. See also... Figure 4This application represents the electrical response observed at each common drive frequency as a combination of direction-independent and direction-dependent response components. Generally, it can be expressed as:

[0094] in, The electrical response observed at the corresponding trajectory point, For direction-independent response components that are independent of the resonant capture direction, Features that vary with frequency The direction of change is related to the response component.

[0095] For the same common driving frequency The electrical responses originating from the first and second resonant capture trajectories share the same The direction-dependent response components vary with their respective frequency characteristics. Therefore, the electrical response can be decomposed using trajectory data from two different capture directions according to a preset response combination relationship. As the transducer system response used for subsequent transducer system parameter identification.

[0096] In a preferred embodiment, when the preset identification frequency range is sufficiently narrow and the direction-dependent dynamics can be described using a local first-order model, the direction-dependent response components are represented as a linear function of the frequency change rate:

[0097] in, These are dynamic response parameters. When the electrical response... When it is complex admittance, It can be a complex parameter, used to simultaneously describe the direction-dependent changes in amplitude and phase as a function of the rate of frequency change.

[0098] For the same common driving frequency The first trajectory data and the second trajectory data respectively satisfy:

[0099] when When this happens, the direction-independent response components can be obtained:

[0100] And dynamic response parameters:

[0101] Since the preferred first and second trajectory data correspond to opposite directions of frequency change, they can usually satisfy... This avoids the two frequencies changing at excessively similar rates, which could increase the sensitivity of the solution. However, this application does not require that their absolute values ​​be equal.

[0102] When the frequency change rates of the first and second trajectories at the common driving frequency are approximately symmetrical, that is... In this case, the direction-independent response component can also be approximated as the average of two electrical responses. This approximation is only a simplified implementation for a special case and is not intended to define the general response decomposition relationship.

[0103] In addition to the local linear model described above, in other embodiments, the direction-dependent response components... Alternatively, pre-calibrated nonlinear functions, piecewise functions, lookup tables, or parameterized models containing multiple parameters to be identified can be used. For example, directionally correlated model parameters and directionally independent response components can be jointly fitted based on two sets of directional data at multiple common driving frequencies. Therefore, the decomposition of the electrical response according to the combination relationship in this application is not limited to the above-mentioned linear solution form.

[0104] Step S105: Identify transducer system parameters based on the direction-independent response.

[0105] Figure 5 This is a schematic diagram of the equivalent circuit of a parameter model of an ultrasonic transducer system provided in an embodiment of this application. See also... Figure 5 In one implementation, the preset transducer system parameter model adopts a Butterworth-Van Dyke equivalent model. This model includes equivalent dynamic resistance... Equivalent dynamic inductance and equivalent dynamic capacitance The dynamic branch formed by series connection, and the equivalent parallel parasitic capacitance connected in parallel with the dynamic branch. .

[0106] Let the angular frequency be:

[0107] The complex admittance corresponding to this parameter model can then be expressed as:

[0108] The digital control module 130 obtains direction-independent responses at multiple common drive frequencies:

[0109] and the corresponding angular frequency, for model parameters , , and Perform fitting. For example, the first... The common driving frequency is denoted as The following weighted least squares objective can be solved: Wherein, represents the number of common driving frequencies participating in the transducer system parameter identification, represents the sequence number of the common driving frequencies, and represents the weighting coefficient corresponding to the i-th common driving frequency. Each weighting coefficient can be set to the same value, or it can be determined based on at least one of signal-to-noise ratio, interpolation distance, and data stability.

[0110]

[0111] in, This indicates the number of common drive frequencies involved in the transducer system parameter identification. The sequence number indicating the common drive frequency. Indicates the first The weighting coefficients correspond to the common driving frequencies. Each weighting coefficient can be set to the same value, or it can be determined based on at least one of signal-to-noise ratio, interpolation distance, and data stability. The solution algorithm can employ nonlinear least squares, the Gauss-Newton algorithm, the Levenberg-Marquardt algorithm, recursive parameter estimation, or other algorithms capable of obtaining model parameters from the complex response at multiple frequencies. This application is not limited to a specific numerical optimization algorithm and therefore is not restricted by any particular optimization algorithm.

[0112] In other embodiments, the preset transducer system parameter model can also be other equivalent parameter models that can characterize the frequency characteristics of the target resonant mode. When multiple resonant modes exist, a corresponding local parameter model can be set for the target resonant mode to be controlled; this embodiment does not require that all resonant modes be identified simultaneously.

[0113] Besides the electromechanical parameters of the transducer system itself, the relative gain and relative time delay between the voltage detection link and the current detection link may also affect the measured complex admittance and phase relationship. Therefore, in a further embodiment, the detection link parameters can be identified based on at least one corresponding voltage response and current response during the first and second resonant capture processes.

[0114] definition This is the real gain factor in the relative complex amplitude calibration of the current sensing link relative to the voltage sensing link. This represents the signed relative time delay of the current sensing link compared to the voltage sensing link. When the current sensing link has a delay relative to the voltage sensing link... When the complex admittance is measured, it can be expressed as:

[0115] in, The admittance of the transducer system is given without considering the relative gain and relative time delay of the detection link.

[0116] The detection link parameters can be pre-identified using known reference states, or they can be jointly fitted with the transducer system parameters. For example, the following can be constructed:

[0117] The set of parameters to be identified is as follows:

[0118] Solve based on the direction-independent response of multiple common driving frequencies:

[0119] This allows for the simultaneous acquisition of transducer system parameters and detection link parameters when the identification conditions are met. The identification conditions may include the number of common drive frequencies and frequency coverage that meet the requirements for the number of parameters to be identified and parameter sensitivity. When the preset identification conditions are not met, some detection link parameters determined by factory calibration, historical calibration, or other reference states can be fixed, and only the remaining parameters to be identified can be updated. If some detection link parameters have already been determined by factory calibration, the known parameters can also be fixed, and only the remaining parameters can be updated.

[0120] Step S106: Perform self-calibration of digital ultrasound generator parameters based on the identified parameters.

[0121] After obtaining the transducer system parameters and / or detection link parameters, the digital control module 130 can configure at least one of the resonance tracking control parameters and ultrasonic output control parameters of the digital ultrasonic generator 100.

[0122] First, it can be based on the equivalent parallel parasitic capacitance. Parasitic current compensation is performed on the current response used for resonant trapping. The current component generated by the parasitic capacitance can be expressed as:

[0123] From this, we can obtain the compensation current used to characterize the dynamic branch:

[0124] in, This is the total current complex response obtained by the current detection unit 122. Subsequent phase detection can use... and The phase relationship between them is determined to reduce the influence of parallel parasitic capacitance on resonance determination.

[0125] Second, it can be based on the relative delay parameters of the detection link. Correct the phase relationship between the voltage response and the current response. When When defined as the delay of the current sensing link relative to the voltage sensing link, it can be used as follows:

[0126] The phase difference after detecting link delay correction is obtained. Among them, The measured phase difference is determined based on the complex responses of the original voltage and current. This is the corrected phase difference.

[0127] Third, it can be based on the equivalent dynamic inductance. and equivalent dynamic capacitance Determine the estimated series resonant frequency of the target mode:

[0128] Can This serves as the basis for selecting the initial frequency for subsequent resonance capture. For example, during the next startup, the initial drive frequency is set to be within the range specified in the original text. The nearby preset offset position reduces the search range for invalid frequencies.

[0129] Fourth, it can be based on , , The resonant bandwidth, damping characteristics, or phase slope near resonance obtained from these parameters are used to configure the resonant tracking control parameters. Specifically, the control parameters can be calculated analytically, or a mapping table between the transducer system parameters and the controller's proportional, integral, frequency limiting, or frequency step size can be pre-established, and the corresponding parameters can be selected based on the identification results. Since the closed-loop control structure can differ between different digital ultrasonic generators, this application does not require a single formula for calculating the control parameters.

[0130] Fifth, ultrasonic output control parameters can be configured based on the impedance characteristics represented by the transducer system parameters, such as output current setpoint, output voltage limit, amplitude adjustment range, power limit, or ultrasonic output reference for different bonding stages. Through the above configuration, the output control of the digital ultrasonic generator can be matched with the actual electrical characteristics of the current ultrasonic transducer system.

[0131] Therefore, this embodiment does not directly use the dynamic electrical response under a single capture direction as the response characterizing the frequency characteristics of the transducer system. Instead, it utilizes the response difference provided by two opposite capture directions to separate the direction-dependent response components, and then uses the direction-independent response for parameter identification. This reduces the impact of the resonant capture direction and frequency variation process on the transducer system parameter identification. Furthermore, both sets of identification data originate from the actual resonant capture process of the digital ultrasonic generator, reducing reliance on independent and complete frequency sweep tests.

[0132] Based on the above-described overall method embodiments, this application also provides a second embodiment of trajectory data enhancement based on identified anchor points and short-term dwell time. To improve the measurement stability of the electrical response of some trajectory points, identified anchor points can also be set during the formation of at least one of the first and second resonant capture trajectories.

[0133] The digital control module 130 determines the phase error based on the phase difference between the voltage response and the current response. When the phase error reaches a preset phase error value or enters a preset phase error range, the current trajectory point is designated as an identification anchor point. Different identification anchor points can correspond to different phase states, thus distributing the identification anchor points at different frequency positions near the target resonant mode.

[0134] For anchor point identification, the digital control module 130 temporarily maintains the current driving frequency unchanged and continues for a preset number of driving cycles, synchronously sampling the voltage and current signals during the holding period. The voltage and current complex responses from multiple driving cycles can be averaged or coherently accumulated to update the electrical response corresponding to the identified anchor point. After the holding period ends, the original resonant capture control is restored, and trajectory data acquisition continues along the current direction.

[0135] Since the frequency change rate is zero during the short-term dwell period, the dwell data can be used to improve the stability of the parameter model fitting of the transducer system, verify the rationality of the direction-independent response, or update the electrical response of the trajectory point. For the linear decomposition of the direction-independent response component that relies on two opposite non-zero frequency change rates in the above overall method embodiment, the trajectory data that is in a frequency change state before and after the dwell period can be used for the solution first, instead of requiring the dwell data with a zero frequency change rate to be directly used as one of the two samples for the paired solution.

[0136] The number and phase state of the identification anchor points can be determined based on the width of the identification frequency range, the required data density, and the allowable self-calibration time. Short-term dwell only occurs at a small number of state points during the resonance capture process and does not require dwelling at every point throughout the entire identification frequency range. Therefore, it is still possible to avoid turning the parameter self-calibration process into a complete fixed-frequency sweep test.

[0137] See Figure 2 In a digital ultrasonic generator 100 with parameter self-calibration function, the ultrasonic drive module 110 outputs an ultrasonic drive signal to the ultrasonic transducer system 200; the electrical response acquisition module 120 acquires the voltage response and current response of the ultrasonic transducer system 200; the digital control module 130 determines the electrical response based on the voltage response and current response, and executes the parameter self-calibration method described in the above general method embodiment and / or the second embodiment.

[0138] The direct digital frequency synthesis unit 111 can be implemented by the logic circuitry within the digital control module 130, or it can be configured as an independent digital frequency synthesizer. The digital control module 130 adjusts the driving frequency by changing the frequency control word and can record the driving frequency sequence corresponding to the actual frequency control word during the resonance capture process. Therefore, the frequency change rate can be calculated based on the recorded actual driving frequency sequence or directly determined based on the frequency control quantity within the digital controller.

[0139] The digital-to-analog converter 112 is used to convert the digital drive waveform into an analog waveform; the amplitude adjustment unit 113 can be a multiplicative digital-to-analog converter, a programmable gain amplifier, or other amplitude control circuits; the power amplification unit 114 amplifies the amplitude-adjusted drive signal to a voltage and current level that can drive the ultrasonic transducer system 200.

[0140] The voltage detection unit 121 and the current detection unit 122 each include detection and conditioning circuits adapted to the corresponding signal amplitudes. The analog-to-digital conversion unit 123 can be a single multi-channel analog-to-digital converter or multiple analog-to-digital converters synchronized with each other. The digital control module 130 generates a synchronous sampling timing sequence according to the current driving frequency, so that the voltage and current sampling have a definite phase relationship with the driving cycle.

[0141] Parameter self-calibration can be triggered after the digital ultrasonic generator 100 is powered on, after the ultrasonic transducer 210 or bonding tool 220 is replaced, after a change in resonant frequency exceeding a threshold is detected, after equipment maintenance, or according to a preset cycle. When a complete self-calibration is not required, routine resonance tracking can be performed directly using the transducer system parameters and detection link parameters obtained from the most recent self-calibration.

[0142] See Figure 2 The wire bonding apparatus 400 includes a digital ultrasonic generator 100, an ultrasonic transducer system 200, and other mechanical and control mechanisms for carrying or positioning the bonding object 300. The ultrasonic transducer 210 in the ultrasonic transducer system 200 converts the electrical drive signal output from the digital ultrasonic generator 100 into mechanical ultrasonic vibration, which is then transmitted to the bonding position via the bonding tool 220.

[0143] Before wire bonding or when a preset self-calibration trigger condition is met, the digital ultrasound generator 100 performs bidirectional resonant capture self-calibration. The first resonant capture begins from one side of the reference resonant frequency, and the second resonant capture begins from the other side, acquiring trajectory data within a preset identification frequency range in opposite directions under parameter self-calibration conditions. The digital ultrasound generator 100 obtains the direction-independent transducer system response based on the electrical response and frequency variation characteristics in the two trajectories, thereby identifying the transducer system parameters and configuring the resonant tracking parameters and / or ultrasound output parameters used in the subsequent wire bonding process based on the identification results.

[0144] In one implementation, parameter self-calibration can be triggered when the bonding tool 220 is replaced, the ultrasonic transducer 210 is reinstalled, or the connecting cable is changed. Since these changes may cause variations in resonant frequency, impedance characteristics, parasitic capacitance, or phase relationship, re-identifying the parameters and updating the control configuration of the digital ultrasonic generator 100 helps improve the consistency of ultrasonic drive under different transducer system connection states.

[0145] In actual wire bonding, after completing parameter self-calibration, the system can exit the trajectory acquisition state and maintain the target resonant mode using normal resonant tracking. In other words, the controlled crossing of the preset identification frequency range is mainly used in the self-calibration phase and is not required to be repeated during each actual bonding ultrasound output.

[0146] The above embodiment is illustrated using the example of a first initial driving frequency located on the low-frequency side of the reference resonant frequency and a second initial driving frequency located on the high-frequency side. When the first initial driving frequency is located on the high-frequency side, the first and second acquisition directions can be interchanged, as long as the two resonant acquisitions start from different sides of the reference resonant frequency and have different frequency change directions in the identification data used for response decomposition.

[0147] The first and second initial driving frequencies are not required to be symmetrical about the reference resonant frequency; the first and second resonant capture trajectories are also not required to have exactly the same number of trajectory points, sampling times, or frequency change rates. By constructing and interpolating the common driving frequency, two irregular closed-loop capture trajectories can be converted into paired data suitable for response decomposition.

[0148] The preset identification frequency range described in this application is preferably set around the target resonant mode, but its geometric center is not required to completely coincide with the reference resonant frequency. As long as both resonant capture trajectories cover multiple common driving frequencies used to form paired data, and direction-independent response components can be determined using data from different capture directions, the response decomposition and parameter identification of this application can be achieved.

[0149] The electrical response described in this application is not limited to complex admittance; it can also employ complex impedance or other frequency response quantities determined by voltage and current responses. When using complex impedance, direction-independent impedance components and direction-dependent impedance components corresponding to the complex admittance form can be established, and transducer system parameters can be identified based on the direction-independent impedance at multiple common driving frequencies.

[0150] The transducer system parameter model described in this application is not limited to... Figure 5 The diagram shows a single equivalent circuit form. Figure 5The dynamic and parasitic branches shown are preferred parameter models that are easy to explain; other equivalent parameter models can be selected based on the transducer system structure and the target mode, provided that the frequency characteristics of the target resonant mode can be characterized.

[0151] The specific control thresholds, number of sampling points, identification frequency range width, controller parameters, and parameter fitting algorithms mentioned above can all be set according to the operating frequency, power level, and dynamic characteristics of the specific digital ultrasonic generator and ultrasonic transducer system. Unless expressly defined in the claims, these specific implementation parameters do not constitute a limitation on the scope of protection of this application.

[0152] The above embodiments are for illustrative purposes only and are not intended to limit the scope of this application. For those skilled in the art, any equivalent substitutions, simple modifications, or improvements made without departing from the technical concept of this application should be included within the scope of protection of this application.

Claims

1. A parameter self-calibration method applied to a digital ultrasound generator, characterized in that, include: The digital ultrasonic generator is controlled to resonate from the first initial driving frequency to the resonant state of the ultrasonic transducer system to capture resonance, and the first resonance capture trajectory is obtained. A reference resonant frequency is determined based on the first resonant capture trajectory, and a second initial driving frequency is determined such that the first initial driving frequency and the second initial driving frequency are located on both sides of the reference resonant frequency; resonant capture is performed from the second initial driving frequency to the resonant state to obtain a second resonant capture trajectory with a different capture direction than the first resonant capture trajectory; Based on the electrical responses in the first and second resonant capture trajectories and the capture direction difference between the two resonant capture trajectories, the transducer system response independent of the resonant capture direction is determined. The transducer system parameters are identified based on the transducer system response and the preset transducer system parameter model, and the digital ultrasound generator is self-calibrated based on the transducer system parameters.

2. The parameter self-calibration method according to claim 1, characterized in that, The control of the digital ultrasound generator to perform resonance capture includes: The voltage and current responses of the ultrasonic transducer system are obtained, the phase error is determined based on the phase difference between the voltage and current responses, and the driving frequency of the digital ultrasonic generator is adjusted in a closed loop based on the phase error so that the driving frequency converges to the resonant state of the ultrasonic transducer system. Both the first resonant capture trajectory and the second resonant capture trajectory are formed by the actual closed-loop adjustment process of the driving frequency.

3. The parameter self-calibration method according to claim 2, characterized in that, The first resonant capture trajectory and the second resonant capture trajectory each include multiple trajectory points, and each trajectory point includes a driving frequency, a frequency change feature characterizing the driving frequency as a function of time, and a corresponding electrical response; The determination of the transducer system response independent of the resonant capture direction includes: Based on the driving frequency, frequency change characteristics, and electrical response corresponding to each trajectory point in the first and second resonant capture trajectories, a trajectory data association is established between the two resonant capture trajectories. The response components with consistency under different capture directions are determined according to the trajectory data association, which serve as the transducer system response.

4. The parameter self-calibration method according to claim 3, characterized in that, The frequency change characteristic includes the frequency change rate; A preset identification frequency range is determined near the reference resonant frequency; During the first and second resonant capture processes, the driving frequency is controlled to cross the preset identification frequency range along opposite frequency change directions, so that both the first and second resonant capture trajectories cover the preset frequency range. After the first and second resonance capture processes have completed crossing the preset identification frequency range, the driving frequency is controlled to return to the reference resonance frequency to enter the resonance state.

5. The parameter self-calibration method according to claim 4, characterized in that, Multiple common driving frequencies are determined within the preset identification frequency range; For each of the aforementioned common driving frequencies, the corresponding trajectory data is determined from the first resonant capture trajectory and the second resonant capture trajectory, respectively; When there is a trajectory point in the resonant capture trajectory that has the same common driving frequency, the trajectory data corresponding to the common driving frequency is determined according to the frequency change rate and electrical response corresponding to the trajectory point. When there is no trajectory point in the resonant capture trajectory that is the same as the common driving frequency, interpolation is performed based on the driving frequency, frequency change rate, and electrical response corresponding to the adjacent trajectory points located on both sides of the common driving frequency to obtain the frequency change rate and electrical response at the common driving frequency.

6. The parameter self-calibration method according to claim 5, characterized in that, The electrical responses originating from the first resonant capture trajectory and the second resonant capture trajectory at each of the common driving frequencies are respectively expressed as a combination of a direction-independent response component that corresponds to the current common driving frequency and is independent of the resonant capture direction and a direction-dependent response component that changes with the corresponding frequency. The direction-independent response component is used as the transducer system response; Based on the frequency variation characteristics and electrical responses of the first and second resonant capture trajectories respectively under the same common driving frequency, the electrical response is decomposed according to the combination relationship to determine the direction-independent response component and the direction-dependent response component.

7. The parameter self-calibration method according to claim 6, characterized in that, The direction-dependent response component has a preset linear relationship with the frequency change rate; For the first trajectory data and the second trajectory data respectively derived from the first resonant capture trajectory and the second resonant capture trajectory under the same common driving frequency, the direction-independent response component and the dynamic response parameter characterizing the degree of change of the direction-dependent response component with the frequency change rate are solved according to the electrical response and frequency change rate corresponding to the first trajectory data and the second trajectory data respectively, as well as the preset linear relationship.

8. The parameter self-calibration method according to claim 3, characterized in that, During the formation of at least one of the first resonant capture trajectory and the second resonant capture trajectory, the phase error is determined based on the phase difference between the voltage response and the current response; When the phase error reaches a preset phase error value or enters a preset phase error range, the corresponding trajectory point is determined as the identification anchor point. For the identified anchor point, the current driving frequency is maintained for a preset number of driving cycles, and the electrical response corresponding to the identified anchor point is updated based on the voltage and current responses obtained during the period of maintaining the driving frequency, and then the resonance capture continues.

9. The parameter self-calibration method according to claim 2, characterized in that, The voltage response and the current response are obtained through a sampling process synchronized with the current driving frequency; The voltage and current signals of the ultrasonic transducer system are sampled during at least one complete driving cycle, and the obtained voltage and current sampled values ​​are processed in the frequency domain to determine the voltage complex response and current complex response at the corresponding driving frequency. The electrical response at the corresponding driving frequency is determined based on the voltage complex response and the current complex response, wherein the electrical response includes at least one of complex impedance and complex admittance.

10. The parameter self-calibration method according to claim 1, characterized in that, The transducer system parameter model includes a dynamic branch for characterizing the electromechanical dynamic characteristics of the ultrasonic transducer system and a parasitic branch connected in parallel with the dynamic branch. The transducer system parameters include the equivalent dynamic resistance, equivalent dynamic inductance, and equivalent dynamic capacitance corresponding to the dynamic branch, and also include the equivalent parallel parasitic capacitance corresponding to the parasitic branch. The parameter self-calibration method further includes: during the formation of at least one of the first resonant capture trajectory and the second resonant capture trajectory, acquiring the voltage response and current response of the ultrasonic transducer system, and identifying detection link parameters for characterizing the response difference between the voltage detection link and the current detection link based on the voltage response and the current response; The detection link parameters include at least one of the relative gain parameter and the relative delay parameter between the voltage detection link and the current detection link.

11. The parameter self-calibration method according to claim 10, characterized in that, The parameter self-calibration includes at least one of the following: The parasitic current component is determined based on the identified equivalent parallel parasitic capacitance, and the current response used for resonance capture is compensated based on the parasitic current component. Once the relative time delay parameter is identified, the phase relationship between the voltage response and the current response is corrected based on the relative time delay parameter. The estimated resonant frequency of the ultrasonic transducer system is determined based on the equivalent dynamic inductance and the equivalent dynamic capacitance, and the initial drive frequency for the subsequent resonant capture process is configured based on the estimated resonant frequency. The resonant tracking control parameters for closed-loop adjustment of the driving frequency are configured according to at least one of the equivalent dynamic resistance, the equivalent dynamic inductance, and the equivalent dynamic capacitance. Configure the ultrasonic output control parameters of the digital ultrasonic generator according to the impedance characteristics characterized by the transducer system parameters.

12. A digital ultrasonic generator with parameter self-calibration function, characterized in that, It includes an ultrasonic drive module, an electrical response acquisition module, and a digital control module; The ultrasonic drive module is used to output ultrasonic drive signals to the ultrasonic transducer system; The electrical response acquisition module is used to acquire the voltage and current responses of the ultrasonic transducer system. The digital control module is used to determine the electrical response of the ultrasonic transducer system based on the voltage response and the current response, and is used to: The ultrasonic drive module is controlled to resonate from the first initial drive frequency to the resonant state of the ultrasonic transducer system to capture resonance, and the first resonance capture trajectory is obtained. A reference resonant frequency is determined based on the first resonant capture trajectory, and a second initial driving frequency is determined, such that the first initial driving frequency and the second initial driving frequency are located on both sides of the reference resonant frequency; the ultrasonic driving module is controlled to perform resonant capture from the second initial driving frequency to the resonant state, so as to obtain a second resonant capture trajectory with a different capture direction than the first resonant capture trajectory; Based on the electrical responses in the first and second resonant capture trajectories and the capture direction difference between the two resonant capture trajectories, the transducer system response independent of the resonant capture direction is determined. The transducer system parameters are identified based on the transducer system response and the preset transducer system parameter model, and the digital ultrasound generator is self-calibrated based on the transducer system parameters.

13. The digital ultrasound generator according to claim 12, characterized in that, The ultrasonic drive module includes a direct digital frequency synthesis unit, a digital-to-analog conversion unit, an amplitude adjustment unit, and a power amplification unit arranged in sequence. The direct digital frequency synthesis unit is used to generate a frequency-adjustable digital drive signal. The digital drive signal is processed by the digital-to-analog conversion unit, the amplitude adjustment unit, and the power amplification unit and then output to the ultrasonic transducer system. The electrical response acquisition module includes a voltage detection unit, a current detection unit, and an analog-to-digital conversion unit. The voltage detection unit and the current detection unit are used to acquire the voltage signal and the current signal of the ultrasonic transducer system, respectively. The digital control module is used to control the analog-to-digital conversion unit to synchronously sample the voltage signal and the current signal according to the current driving frequency, and to determine the voltage response and the current response based on the synchronous sampling results.

14. A wire bonding apparatus, characterized in that, It includes a digital ultrasound generator and an ultrasound transducer system connected to the digital ultrasound generator; The ultrasonic transducer system includes an ultrasonic transducer and a bonding tool mechanically connected to the ultrasonic transducer. The digital ultrasound generator is used to resonate and capture the ultrasound transducer system from a first initial driving frequency to obtain a first resonant capture trajectory, determine a reference resonant frequency based on the first resonant capture trajectory, and determine a second initial driving frequency, such that the first initial driving frequency and the second initial driving frequency are located on both sides of the reference resonant frequency; and resonate and capture the ultrasound transducer system from the second initial driving frequency to obtain a second resonant capture trajectory with a different capture direction than the first resonant capture trajectory. The digital ultrasonic generator is also used to acquire the voltage and current responses of the ultrasonic transducer system during the first and second resonant capture processes, and to determine the corresponding electrical responses based on the voltage and current responses. The digital ultrasonic generator is also used to determine a transducer system response independent of the resonant capture direction based on the electrical responses in the first and second resonant capture trajectories and the difference in capture direction between the two resonant capture trajectories, to identify transducer system parameters based on the transducer system response and a preset transducer system parameter model, and to configure the ultrasonic control parameters of the digital ultrasonic generator during wire bonding based on the transducer system parameters.

Images