A manufacturing process for HDI boards used in automotive cameras

Abstract

This invention discloses a manufacturing process for HDI boards used in automotive cameras. The process includes: perturbing the interface between the metallized hole walls and the insulating medium in the HDI board with a short-pulse beam to obtain stress resonance perturbation wave field data; recording phase perturbation through a high-pass filter layer to obtain time-series data on stress wave propagation and residence states; extracting spectral energy distribution parameters through two-dimensional Fourier transform to obtain an unsteady-state perturbation fingerprint; applying periodically gated pulse current to the micro-hole region to obtain charge hysteresis response data; applying scanning acoustic excitation and acquiring acoustic wave reflection data to obtain acoustic pressure transition parameters; and performing time synchronization and energy density comparison on the stress spectrum, hysteresis response spectrum, and acoustic pressure transition spectrum to output a stress fluctuation risk assessment result. This invention can detect and predict the dynamic stress fluctuation process of the submicron-level metallized hole wall film of the HDI board under complex loads, establish a quantitative relationship between stress fluctuation modes and potential failure risks, and improve the reliability of automotive camera HDI boards.

Description

Technical Field

[0001] This invention relates to the field of automotive camera HDI board testing technology, and in particular to a manufacturing process for automotive camera HDI boards. Background Technology

[0002] High-density interconnect (HDI) boards are indispensable core circuit carriers in current automotive cameras. They are characterized by employing micro-blind via technology to achieve inter-layer interconnection, significantly improving wiring density and signal integrity. Automotive camera HDI boards differ from ordinary HDI boards; they must maintain stable operation under extreme temperature cycling from -40°C to 125°C, continuous vibration and shock, and high humidity and high salt environments, while also ensuring the integrity of micron-level alignment of the optical system and high-speed image signal transmission. The HDI board design, characterized by high reliability, high density, high performance, and high stability, plays a decisive role in the core functions and overall performance of automotive cameras.

[0003] In existing technologies, the inspection of HDI boards for automotive cameras mainly relies on methods such as X-ray imaging, infrared thermography, optical microscopy, and electrical parameter testing. These methods can effectively detect macroscopic structural defects and electrical connection anomalies. However, with the continuous improvement of automotive camera resolution and the development of autonomous driving technology, some problems in HDI boards are becoming increasingly prominent, namely the stress problem of the metallized hole wall film (referring to the submicron-thick metal conductive layer deposited on the inner wall of the tiny blind buried holes in the HDI board). Under the alternating effects of high-frequency vibration and extreme temperature in vehicles, a complex three-dimensional stress field distribution will be generated inside this film. This stress distribution is in a dynamic fluctuation state and has significant non-uniformity at the submicron scale (1 micrometer = 1 / 1000 millimeter), which makes it impossible for traditional inspection methods to capture this dynamic stress fluctuation process in real time. When stress fluctuations accumulate at the interface between the metallized layer and the insulating layer (the dielectric material used to isolate different conductive layers), it may lead to the formation and propagation of microcracks. These microscopic damages are often difficult to detect in a timely manner, potentially leading to intermittent connection failures (i.e., intermittent conductive paths) or decreased signal integrity (i.e., reduced signal transmission quality) in HDI boards during actual use, severely impacting the reliability of automotive cameras and the safety of autonomous driving systems. Therefore, accurately detecting and predicting the dynamic stress fluctuation evolution of the submicron-level metallized hole wall film of HDI boards under complex environmental loading conditions, thereby preventing potential connection failures, has become an urgent technical problem to be solved. Summary of the Invention

[0004] The main objective of this invention is to solve the technical problem that existing vehicle camera HDI board detection methods have insufficient real-time capture capability in detecting dynamic stress fluctuations in the thin film of submicron-level metallized hole walls.

[0005] The first aspect of the present invention provides a manufacturing process for an HDI board for an automotive camera, the manufacturing process for the HDI board for an automotive camera includes:

[0006] Short-pulse beam perturbation was applied to the interface between the metallized hole wall and the insulating medium in the HDI board to obtain stress resonance perturbation wave field data centered on the hole wall;

[0007] Based on the stress resonance perturbation wave field data, phase perturbation is recorded through a high-pass filter layer to obtain time-series data on the propagation and residence state of stress waves within the submicron structure.

[0008] A two-dimensional Fourier transform is performed on the time-series data to extract the spectral energy distribution parameters. The non-steady-state disturbance fingerprint is obtained by calculating the difference function through multi-point detection.

[0009] Based on the unsteady-state perturbation fingerprint, a periodic gated pulse current is applied to the micropore region, the resistance change trajectory is collected, and charge hysteresis response data is obtained.

[0010] Based on the charge hysteresis response data, a scanning acoustic excitation is applied to the HDI board identification area, and acoustic wave reflection intensity and phase delay data are collected to obtain acoustic pressure transition parameters.

[0011] Based on the acoustic pressure transition parameters, the stress spectrum, hysteresis response spectrum, and acoustic pressure transition spectrum are synchronized in time and compared with energy density. The stress fluctuation risk assessment result of the micropores of the HDI plate is obtained by calculating the fluctuation spectrum coupling factor.

[0012] Preferably, the step of perturbing the interface between the metallized via wall and the insulating medium in the HDI board with a short pulse beam to obtain stress resonance perturbation wavefield data centered on the via wall includes:

[0013] The geometric contour of the micro-blind buried vias in the HDI board is scanned to obtain the distribution and orientation data of the metal grain boundaries on the via walls;

[0014] Based on the metal grain boundary distribution and orientation data, the microscopic energy distribution data of the hole wall surface is obtained by calculating the grain boundary energy barrier.

[0015] Based on the microscopic energy distribution data, a picosecond-level short pulse beam is applied to the region where the energy barrier gradient exceeds a first preset threshold to obtain local energy excitation intensity data.

[0016] Based on the local energy excitation intensity data, the energy conversion coefficient of the interface between the metal grain boundary and the insulating medium is calculated to obtain the interface stress expansion and contraction imbalance data.

[0017] Based on the interface stress expansion and contraction imbalance data and the energy conversion coefficient, stress resonance disturbance wave field data centered on the hole wall is obtained.

[0018] Preferably, the step of obtaining the microscopic energy distribution data of the hole wall surface by calculating the grain boundary energy barrier based on the metal grain boundary distribution and orientation data includes:

[0019] Based on the metal grain boundary distribution and orientation data, a temperature-stress stratification analysis was established along the depth direction of the micro-blind buried hole. Combined with the interface stress state under extreme temperature cycling, grain boundary temperature-stress coupling data was obtained.

[0020] Temperature-stress-displacement field calculations were performed on the interface between the metallized layer and the insulating medium in the grain boundary temperature-stress coupling data. Combined with the gradient distribution of the thermal expansion coefficient of the metal interconnect layer, the micro-strain energy distribution data of the interface was obtained.

[0021] Based on the micro-strain energy distribution data of the interface, the dislocation climb and slip critical energies of the metallization layer grain boundaries under thermal cycling load are calculated to obtain the grain boundary barrier distribution map data.

[0022] Based on the grain boundary barrier distribution data, combined with the heat flux density distribution and stress field intensity at the hole wall, the microscopic energy distribution data of the hole wall surface is obtained.

[0023] Preferably, the step of obtaining time-series data on the propagation and residence states of stress waves within the submicron structure by performing phase perturbation recording through a high-pass filter layer based on the stress resonance perturbation wave field data includes:

[0024] Optical path difference analysis was performed on the stress resonance disturbance wave field data to obtain the photoelastic property change path data;

[0025] Based on the photoelastic property change path data, the incident beam is deflected and separated by a high-pass filter layer to obtain phase perturbation intensity distribution data;

[0026] Based on the phase perturbation intensity distribution data, interferometric analysis is performed on the optical path difference variation within the submicron structure of the metallized hole wall to obtain stress wave propagation path data.

[0027] Based on the stress wave propagation path data, differential calculation is performed on the optical path difference phase change of the detection points at preset distances to obtain the stress wave residence location data.

[0028] Based on the stress wave residence location data and stress wave propagation path data, the phase change sequence of each detection point is sampled and analyzed at time intervals to obtain the time series data of the stress wave propagation and residence state within the submicron structure.

[0029] Preferably, the step of performing a two-dimensional Fourier transform on the time-series data to extract spectral energy distribution parameters, and calculating the non-steady-state perturbation fingerprint through a difference function of multi-point detection, includes:

[0030] The time-series data is collected in segments according to multiple detection points along the circumferential direction of the hole wall. Combined with the grain boundary morphology characteristics of the metallized layer surface, spatial sampling sequence data related to the surface microstructure is obtained.

[0031] Based on the spatial sampling sequence data, a two-dimensional Fourier transform is performed on the time-series signal of each detection point, and the energy distribution of the metallization layer grain boundary is mapped to obtain the frequency domain main peak and grain boundary coupling distribution data.

[0032] Based on the frequency domain main peak and grain boundary coupling distribution data, the spectral tail coefficient, energy cross modulation ratio and grain boundary stress fluctuation coefficient of each detection point are extracted to obtain spectral distortion characteristic data.

[0033] By comparing the difference in spectral parameters of adjacent detection points in the spectral distortion feature data, and combining the main peak displacement, energy diffusion rate and grain boundary stress fluctuation coefficient, an unsteady-state disturbance fingerprint is obtained.

[0034] Preferably, the step of comparing the difference in spectral parameters of adjacent detection points in the spectral distortion feature data, and combining the main peak displacement, energy diffusivity, and grain boundary stress fluctuation coefficient to obtain the unsteady-state perturbation fingerprint includes:

[0035] Optical transmittance response analysis was performed on the spectral distortion characteristic data, and stress-optical coupling characteristic data was obtained by combining the photoelectric polarization effect of the metallization layer.

[0036] Based on the stress-optical coupling characteristic data, the optical path difference variation of the micro-aperture array and the imaging sharpness attenuation coefficient are calculated, and combined with the signal transmission bandwidth requirements, photoelectric signal integrity data are obtained.

[0037] Quantum tunneling conductivity analysis is performed on regions in the photoelectric signal integrity data where the signal-to-noise ratio is lower than a second preset threshold. Combined with the interface electronic state density distribution, interface quantum transport characteristic data is obtained.

[0038] Based on the interface quantum transport characteristic data, combined with the main peak displacement, energy diffusion rate and the grain boundary stress fluctuation coefficient, an unsteady-state perturbation fingerprint is obtained.

[0039] Preferably, the step of applying a periodic gated pulse current to the micropore region based on the unsteady-state perturbation fingerprint, acquiring the resistance change trajectory, and obtaining charge hysteresis response data includes:

[0040] Based on the unsteady-state perturbation fingerprint, the stress concentration point distribution in the micropore region is extracted to obtain the electron transport channel data of the metallization layer;

[0041] The electron transport channel is partitioned and mapped according to the stress gradient direction, and a frequency-modulated gated pulse current is applied to obtain charge migration polarization intensity data.

[0042] Based on the charge migration polarization intensity data, the modulation effect of the metal grain boundary stress field on electron scattering is calculated, and the scattering modulation coefficient data is obtained.

[0043] Based on the scattering modulation coefficient data, the quantum tunneling probability and interface barrier height during electron transport are calculated to obtain quantum transport characteristic data;

[0044] Based on the quantum transport characteristic data, the coupling effect of electron scattering and quantum tunneling is calculated to obtain charge hysteresis response data.

[0045] Preferably, the step of applying scanning acoustic excitation to the HDI board identification area based on the charge hysteresis response data, acquiring acoustic wave reflection intensity and phase delay data, and obtaining acoustic pressure transition parameters includes:

[0046] Based on the charge hysteresis response data, the acoustic impedance of the dislocation slip band distribution in the metallized hole wall structure of the HDI plate is calculated to obtain the acoustic impedance distribution data of micro-defects.

[0047] The scanning area is determined based on the micro-defect acoustic impedance distribution data. A scanning acoustic excitation with a frequency range of 0.5 to 3 MHz is applied. Based on the acoustic wave coupling mechanism between the metal grain boundary and the insulating medium interface, acoustic wave energy conversion data is obtained.

[0048] Based on the acoustic energy conversion data, the modulation effect of the metallization layer lattice distortion on the acoustic dispersion effect is calculated to obtain acoustic dispersion characteristic data.

[0049] The acoustic phase transition is calculated from the acoustic dispersion characteristic data, and the lattice distortion acoustoelastic response data is obtained by combining the dislocation density and acoustic wave interaction.

[0050] Based on the lattice distortion acoustic response data, the coupling parameters between microscopic acoustic wave scattering and grain boundary dislocations are extracted to obtain the acoustic pressure transition parameters.

[0051] Preferably, the step of performing nonlinear acoustoelastic analysis based on the acoustic wave dispersion characteristic data, and combining the dislocation density with the acoustic wave interaction to obtain lattice distortion acoustoelastic response data, includes:

[0052] A temperature-velocity correlation mapping is established on the sound wave dispersion feature data, and the temperature-velocity coupling coefficient data is obtained by combining the sound wave propagation characteristics under extreme temperature cycles.

[0053] Based on the temperature-velocity coupling coefficient data, the acoustic phase velocity distortion of the micro-hole array under thermal stress is calculated, and the nonlinear acoustoelastic response data is obtained by combining the interface acoustic impedance jump characteristics.

[0054] The nonlinear acoustoelastic response data are subjected to acoustic wave phase-stress field-temperature field correlation analysis, and combined with the interface stress concentration factor, the interface stress-acoustic wave interaction data are obtained.

[0055] Based on the interface stress-acoustic interaction data, combined with the dislocation density distribution and lattice distortion degree, lattice distortion acoustoelastic response data are obtained.

[0056] Preferably, the step of performing time synchronization and energy density comparison of the stress spectrum, hysteresis response spectrum, and sound pressure transition spectrum based on the acoustic pressure transition parameters, and calculating the stress fluctuation risk assessment result of the HDI plate micropores through the fluctuation spectrum coupling factor, includes:

[0057] Based on the acoustic pressure transition parameters, the stress wave resonance frequency, electron scattering frequency, and acoustic wave scattering frequency of the metallized hole wall grain boundary are phase-locked to obtain critical resonance point distribution data.

[0058] Based on the critical resonance point distribution data, the interaction strength of the dislocation stress field, charge polarization field and acoustic strain field at the metal grain boundary is calculated to obtain multi-field coupling strength data.

[0059] Based on the multi-field coupling strength data, the stress field distortion, carrier concentration gradient and acoustic dispersion coefficient at the grain boundary of the pore wall are extracted to obtain the interface bonding stability data.

[0060] Stress-temperature phase diagram analysis was performed on the micro-defects in the interface stability data, and the interface micro-failure threshold data was obtained by combining the shear slip resistance of lattice dislocations.

[0061] Based on the interface micro-failure threshold data, and combined with the cross-scale response characteristics of hole wall stress resonance, carrier migration and acoustic wave scattering, the stress fluctuation risk assessment results of HDI plate micropores are obtained.

[0062] The manufacturing process provided by this invention addresses the problem of submicron-scale dynamic stress fluctuations in the metallized via walls of automotive camera HDI boards under extreme load conditions. The essence of this problem lies in the fact that, under the alternating effects of thermal cycling and mechanical vibration, the metallized conductive layer deposited on the inner wall of tiny blind vias experiences a non-uniform stress distribution in space, manifesting as a complex dynamic fluctuation process over time. This stress evolution process is located within the structure and has a scale less than 1 micrometer, making it difficult to achieve in-situ tracking with high temporal and spatial resolution using traditional thermal imaging, electrical parameter, or optical observation methods. These problems directly lead to the difficulty in early identification of potential microcracks, resulting in the risk of intermittent connection failures and decreased signal integrity.

[0063] The first step in this manufacturing process is to apply a short-pulse beam perturbation to the interface region between the metallized hole wall and the insulating medium. The perturbation essentially utilizes the difference in thermophysical properties between the metal and the insulating material to create a localized inconsistency in temperature and expansion rate under picosecond-level light pulse irradiation, thereby inducing an imbalance in stress expansion and contraction at the interface. This localized energy perturbation triggers a stress resonance response centered on the hole wall, forming a perturbation wave field that includes reflection, residence, and diffraction effects. The stress wave propagates along the three-dimensional structure of the hole wall, carrying information about the local structural state, thus forming a physically traceable wave unit.

[0064] After the stress-perturbed wavefield is formed, a high-pass filter layer is introduced to record its propagation and residence behavior through phase perturbation. The minute elastic deformation induced by stress causes a change in refractive index, which in turn affects the optical path length of the incident beam, resulting in a measurable phase change. This recording process enables the dynamic capture of the wave state inside the submicron structure under non-contact and non-destructive conditions, and obtains a time-stamped sequence of data to reflect the propagation path, reflection point location, and residence duration of the stress wave.

[0065] After acquiring this set of time-series data, a two-dimensional Fourier transform was performed to extract the energy distribution characteristics in the frequency domain from the time-domain signal. This spectral processing can identify the frequency response characteristics of different locations on the aperture wall under stress disturbance. By analyzing the main peak frequency, spectral tail length, and modulation behavior, spectral parameters characterizing the unstable response are obtained. Through multi-point detection and calculation of the difference function on the spectral characteristics of adjacent detection points, spectral anomaly regions are further screened out, i.e., locations where abnormal accumulation of fluctuation energy or inconsistent modulation occurs, thus forming a "non-steady-state disturbance fingerprint" of stress fluctuations. This fingerprint reflects the non-uniformity of local stress response and is a high-risk sign that may develop into structural damage.

[0066] After identifying the unsteady perturbation region, a periodic gated pulse current is applied to the microporous region. This electrical excitation process induces electrons to migrate within the metallization layer at a predetermined frequency, and the dynamic changes in their resistance trajectories reflect the influence of the material structure on electron transport behavior. In stress concentration regions, grain boundary perturbations or microvoids increase the scattering probability in the electron migration path, resulting in a hysteresis and nonlinear jump in the response to resistance changes. By recording this hysteresis behavior, charge hysteresis response data reflecting the interface state are obtained, capturing the modulation effect of stress on microscopic transport properties from an electrical perspective.

[0067] Subsequently, frequency-scanning acoustic excitation is applied to the identified region. As the sound wave propagates within the microstructure of the pore wall, its propagation speed and reflection intensity are modulated by the local elastic coefficients and density distribution of the material. When stress distribution is uneven or microscopic distortions occur in the crystal structure, the sound wave will experience phase abrupt changes, reflection intensity variations, and sound pressure transitions as it passes through this region. By recording the sound wave propagation behavior at different frequencies and extracting the corresponding sound wave reflection intensity and phase delay characteristics, an acoustic dataset reflecting the integrity of the microstructure can be formed. This data further supplements the information on the mechanical response of the structure under stress disturbance.

[0068] Finally, the information from the three physical dimensions mentioned above is analyzed collaboratively. Data of various types is synchronized in time and aligned in spatial location. For each micropore region, its stress wave response spectrum, charge hysteresis trajectory, and acoustic pressure transition characteristics are compared under the same energy excitation background. By comparing the consistency of energy distribution and evolution trends of each physical response, it is determined whether there are multi-physics coupling anomalies in the region. The coupling strength index formed based on this comparison result serves as the risk assessment basis for the pore wall stress fluctuation state, enabling the identification of potentially high-risk local areas before failure.

[0069] Through this multi-step detection process, the actual stress fluctuation process inside the material no longer relies on reverse analysis after structural failure. Instead, through multi-field collaborative detection with high temporal accuracy and high spatial resolution, accurate identification and classification warnings are achieved before the microscopic response expands into a macroscopic fault, fundamentally improving the ability to deal with reliability failure issues of automotive camera HDI boards. Attached Figure Description

[0070] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0071] Figure 1This is a schematic diagram of one embodiment of the manufacturing process of an HDI board for an automotive camera according to the present invention.

[0072] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0073] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0074] It should be noted that if the embodiments of the present invention involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of the components in a specific posture. If the specific posture changes, the directional indicators will also change accordingly.

[0075] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the term "and / or" throughout the text includes three solutions; taking A and / or B as an example, it includes technical solution A, technical solution B, and a technical solution that simultaneously satisfies A and B. Furthermore, the technical solutions of various embodiments can be combined with each other, provided that they are feasible for those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0076] One embodiment of this application provides a manufacturing process for an HDI board for an automotive camera. Figure 1 This is a flowchart illustrating a manufacturing process for an HDI board for an automotive camera, provided as an embodiment of this application. In this embodiment, the method includes:

[0077] Please see Figure 1 Short pulse beams were used to perturb the interface between the metallized hole wall and the insulating medium in the HDI board to obtain stress resonance perturbation wave field data centered on the hole wall.

[0078] In one embodiment of the present invention, the step of perturbing the interface between the metallized via wall and the insulating medium in the HDI board with a short pulse beam to obtain stress resonance perturbation wavefield data centered on the via wall includes:

[0079] The geometric contour of the micro-blind buried vias in the HDI board is scanned to obtain the distribution and orientation data of the metal grain boundaries on the via walls;

[0080] Based on the metal grain boundary distribution and orientation data, the microscopic energy distribution data of the hole wall surface is obtained by calculating the grain boundary energy barrier.

[0081] Based on the microscopic energy distribution data, a picosecond-level short pulse beam is applied to the region where the energy barrier gradient exceeds a first preset threshold to obtain local energy excitation intensity data.

[0082] Based on the local energy excitation intensity data, the energy conversion coefficient of the interface between the metal grain boundary and the insulating medium is calculated to obtain the interface stress expansion and contraction imbalance data.

[0083] Based on the interface stress expansion and contraction imbalance data and the energy conversion coefficient, stress resonance disturbance wave field data centered on the hole wall is obtained.

[0084] The following is a detailed description of the steps involved in the above embodiments:

[0085] Contour scanning of the geometric profile of micro-blind and buried vias in HDI boards to obtain data on the distribution and orientation of metal grain boundaries in the via walls can be achieved using confocal laser scanning microscopy (CLSM) combined with X-ray diffraction (XRD). First, a high-resolution CLSM is used to create a three-dimensional image from the via opening. Through layer-by-layer optical slicing, the internal contour map of the micro-blind and buried vias is constructed without damaging the HDI board structure. Then, a micro-focused X-ray diffraction system is used to analyze the crystal structure of the target region. The X-ray beam is focused to the micrometer level through a collimation system, penetrating the surface of the HDI board to directly obtain grain boundary information of the internal metallization layer. In practice, the X-ray energy is set in the range of 8-15 keV, and the scanning step size is 1-2 micrometers to ensure sufficient spatial resolution without damaging the board structure. This non-destructive scanning method can accurately characterize the grain boundary distribution and grain orientation features of the metal layer on the hole wall, providing basic data for subsequent analysis, and fully complies with the requirement that the inspection of vehicle camera HDI boards must not damage the original structure.

[0086] Based on the aforementioned metal grain boundary distribution and orientation data, the microscopic energy distribution data of the pore wall surface is obtained through grain boundary energy barrier calculation, achieved using the quantum grain boundary energy analysis method. First, the grain boundary distribution data obtained from X-ray diffraction is converted into a digital mesh, and the Read-Shockley equation is applied to calculate the fundamental energy for different types of grain boundaries. Then, combined with the temperature and stress field conditions of the pore wall under actual working conditions, the grain boundary energy distribution under actual working conditions is calculated using the principles of continuum mechanics. The meshing accuracy during the calculation is 10-20 nanometers to ensure that subtle changes in grain boundary energy are captured. For copper metallization layers commonly found in automotive camera HDI boards, high-angle grain boundaries (mismatch angle greater than 15°) typically exhibit higher energy states, and these regions are more likely to become stress concentration points under extreme temperature cycling conditions from -40°C to 125°C. The effect of this step is to transform static grain boundary structure information into dynamic energy distribution data, particularly identifying high-energy regions and regions with abrupt energy barrier changes in the grain boundary network. These regions are often sensitive points for subsequent stress response and potential failure initiation locations.

[0087] Based on the microscopic energy distribution data, picosecond-level short-pulse beams are applied to regions where the energy barrier gradient exceeds a first preset threshold to obtain local energy excitation intensity data. This is achieved using an ultrafast laser system. First, regions where the energy barrier gradient exceeds a preset threshold (typically set to more than three times the average gradient value) are identified. These regions correspond to abrupt changes in grain boundary types or the junctions of high- and low-angle grain boundaries. Then, a picosecond laser with a wavelength of 532 nm generates laser pulses with a pulse width of 5-20 picoseconds. A precision focusing system focuses the beam to a spot size of 1-2 micrometers to accurately irradiate the target region. The laser repetition frequency is set in the range of 10-50 kHz, and the single-pulse energy is controlled in the microjoule range to ensure that only transient thermal gradients are generated in the target region without causing material damage. Simultaneously, a high-speed infrared thermal imager monitors the temperature change of the irradiated area in real time, and the local energy excitation intensity data is calculated by combining this data with the material's thermophysical parameters. Picosecond pulses were chosen instead of nanosecond or femtosecond pulses because the picosecond timescale matches the phonon relaxation time of the metal lattice, enabling efficient conversion of light energy into heat and mechanical energy, generating controllable stress waves without causing permanent deformation or damage to the material.

[0088] Based on the local energy excitation intensity data, the energy conversion coefficient of the interface between the metal grain boundary and the insulating medium is calculated to obtain the interfacial stress expansion and contraction imbalance data, which is achieved through a thermo-mechanical coupling analysis method. First, a framework for analyzing the dual-material interface structure, including the metallization layer and the insulating medium, is established using thermodynamic principles. The thermal expansion coefficients of the two materials are input (copper is approximately 16.5 × 10⁻⁶). / K, epoxy resin is approximately (60-100) × Physical parameters such as K / m², elastic modulus (approximately 110-130 GPa for copper and 3-4 GPa for epoxy resin), and thermal conductivity (approximately 400 W / m·K for copper and 0.1-0.2 W / m·K for epoxy resin) were calculated. Then, based on the energy intensity data generated by laser excitation, the energy conversion coefficient at the interface was calculated, which characterizes the efficiency of thermal energy conversion to mechanical energy. Combined with the local temperature gradient, the non-uniform expansion of the materials on both sides of the interface was calculated to obtain the interfacial stress expansion and contraction imbalance data. The time step was set to picoseconds, and the spatial resolution to 10-50 nanometers to accurately capture the rapid changes at the interface. Interfacial stress expansion and contraction imbalance is a key indicator for evaluating interfacial bonding strength because shear stress in an unbalanced state is more likely to lead to interfacial separation and microcrack formation, which is particularly critical under the high-frequency vibration conditions of an automotive environment.

[0089] Based on the interfacial stress expansion and contraction imbalance data and the energy conversion coefficient, stress resonance disturbance wave field data centered on the hole wall is obtained through elastic wave propagation analysis and resonance characteristic analysis. First, using the interfacial stress expansion and contraction imbalance data as initial conditions, the stress wave propagation characteristics are calculated using the Navier-Stokes equations in continuum mechanics. Considering the cylindrical geometry of the metal hole wall, the complete wave field distribution, including longitudinal waves, transverse waves, and surface waves, is analyzed. Particular attention is paid to the reflection, refraction, and mode conversion effects of stress waves at the metal-insulator interface, and the energy distribution of different frequency components is calculated. Normalization is performed using the energy conversion coefficient to identify resonance frequency points with high energy density and stable waveforms. The frequency analysis range is set from 1MHz to 100MHz, covering the characteristic vibration frequencies of micron and submicron scale structures. For a typical copper metallized hole wall with a diameter of 80 microns and a wall thickness of 5 microns in a typical automotive camera HDI board, its radial fundamental frequency resonance point is typically in the range of 10-15MHz. Stress wave field data focusing on resonant frequency points are of great significance because materials are most sensitive to the resonant state. Tiny structural defects can cause resonant frequency shifts or resonant peak broadening, providing a highly sensitive means of identifying potential failure points. This is crucial for ensuring the long-term reliability of vehicle cameras under extreme temperature cycling and continuous vibration environments.

[0090] In one embodiment of the present invention, obtaining the microscopic energy distribution data of the hole wall surface by calculating the grain boundary energy barrier based on the metal grain boundary distribution and orientation data includes:

[0091] Based on the metal grain boundary distribution and orientation data, a temperature-stress stratification analysis was established along the depth direction of the micro-blind buried hole. Combined with the interface stress state under extreme temperature cycling, grain boundary temperature-stress coupling data was obtained.

[0092] Temperature-stress-displacement field calculations were performed on the interface between the metallized layer and the insulating medium in the grain boundary temperature-stress coupling data. Combined with the gradient distribution of the thermal expansion coefficient of the metal interconnect layer, the micro-strain energy distribution data of the interface was obtained.

[0093] Based on the micro-strain energy distribution data of the interface, the dislocation climb and slip critical energies of the metallization layer grain boundaries under thermal cycling load are calculated to obtain the grain boundary barrier distribution map data.

[0094] Based on the grain boundary barrier distribution data, combined with the heat flux density distribution and stress field intensity at the hole wall, the microscopic energy distribution data of the hole wall surface is obtained.

[0095] The following is a detailed description of the steps involved in the above embodiments:

[0096] Based on the metal grain boundary distribution and orientation data, a temperature-stress stratification analysis was established along the depth direction of the micro-blind buried vias. Combined with the interfacial stress state under extreme temperature cycling, grain boundary temperature-stress coupling data was obtained through thermal stress field analysis and temperature gradient mapping techniques. First, an infrared thermal imager was used to perform temperature cycling tests on the HDI board within a temperature range of -40°C to 125°C. The dwell time at each temperature point was 30 minutes, and the heating and cooling rate was controlled at 2-3°C / minute to ensure a stable temperature gradient along the board thickness. Then, a fiber Bragg grating (FBG) sensor array was used to set measurement points at 20-micrometer intervals along the via depth direction to collect stress state data at different depths in real time. The collected temperature and stress field data were spatially registered with the previously obtained metal grain boundary distribution data to establish the stress response relationship of each grain boundary region under different temperature conditions, forming a stress-temperature stratification map along the depth direction. For example, in some operating environments of automotive cameras, when the ambient temperature drops sharply from 25°C to -30°C, the 20-30°C temperature difference between the surface and interior of the HDI board can cause a stress gradient of 80-120 MPa at the grain boundaries of the metallization layer. The effect of this analytical step is to reveal how the temperature gradient affects the stress state of the metal grain boundaries along the depth direction of the blind / buried via, providing a quantitative basis for identifying stress-prone areas under temperature cycling loads.

[0097] Temperature-stress-displacement field calculations were performed on the interface between the metallized layer and the insulating medium in the grain boundary temperature-stress coupling data. Combined with the gradient distribution of the thermal expansion coefficient of the metal interconnect layer, the microscopic strain energy distribution data of the interface was obtained. This was achieved through Digital Image Correlation (DIC) and joint thermo-mechanical field calculations. First, a high-resolution DIC system was used to capture the microscopic displacement field of the material interface during temperature cycling, achieving a spatial resolution of 0.1 micrometers and a displacement measurement accuracy of 10 nanometers. Then, the thermal expansion coefficient distribution of the metallized layer in different regions, especially the thermal expansion coefficient gradient near the grain boundaries, was measured using a Thermal Expansion Analyzer. The displacement field data and thermal expansion coefficient data were input into the thermo-mechanical field calculation program to calculate the strain energy distribution at the interface. In the calculation, the local strain energy density E was obtained by integrating the stress tensor σij and the strain tensor εij, taking into account the material anisotropy and the spatial non-uniformity of the thermal expansion coefficient. For example, at the copper-epoxy resin interface, due to the difference in thermal expansion coefficients (copper is 16.5 × 10⁻⁶), the strain energy distribution at the interface was significantly increased. / K, epoxy resin is (60-100)× At 85°C, the strain energy density at the interface can reach 0.5-0.8 J / K. This exceeds the interfacial binding energy in some regions. The effect of this step is that by accurately calculating the microscopic strain energy distribution at the interface, it quantitatively assesses the impact of thermal expansion mismatch on interfacial stability, establishing a physical basis for predicting the risk of interfacial separation.

[0098] Based on the interface micro-strain energy distribution data, the dislocation climb and slip critical energies of the metallization layer grain boundaries under thermal cycling loads were calculated, and grain boundary barrier distribution data were obtained. This was achieved through micromechanical analysis and dislocation dynamics calculations. First, based on the fine structure of the grain boundary region obtained by transmission electron microscopy (TEM), a digital characterization including dislocation density and distribution was established. Then, based on the interface micro-strain energy distribution data, the dislocation climb and slip energies of different types of grain boundaries (such as high-angle grain boundaries, low-angle grain boundaries, twin boundaries, etc.) under thermal cycling loads were calculated. The calculations considered the interactions between dislocations and grain boundaries, between dislocations, and the influence of temperature on dislocation motion, and the climb critical energy Ec and slip critical energy Es were obtained respectively. For copper metallization layers commonly found in HDI boards, the dislocation climb critical energy of high-angle grain boundaries (>15°) is typically 0.7-1.2 eV, and the slip critical energy is 0.3-0.5 eV. These values ​​change with temperature. This step links interfacial strain energy with the microscopic dislocation motion mechanism of grain boundaries, constructs a grain boundary barrier distribution map, reveals the mechanical response characteristics and failure sensitivity of various grain boundaries under thermal cycling conditions, and provides a microscopic mechanism basis for subsequent identification of potential microcrack initiation locations.

[0099] Based on the grain boundary barrier distribution data, combined with the heat flux density distribution and stress field intensity at the hole wall, the microscopic energy distribution data of the hole wall surface is obtained through heat flow field analysis and comprehensive energy assessment methods. First, a heat flux sensor array is used to measure the heat flux density distribution around the hole wall of the HDI board under actual operating conditions. The sensor spacing is 100 micrometers, and the measurement accuracy is 0.01 W / m². Simultaneously, phase-shifting digital holography was used to measure the stress field distribution on the hole wall surface, achieving a stress resolution of 1 MPa. Then, the heat flux density data, stress field data, and grain boundary barrier distribution map were spatially superimposed and analyzed to calculate the comprehensive energy state at each location, forming a microscopic energy distribution map of the hole wall surface. In this analysis, the location energy value E comprehensively considers three components: grain boundary barrier energy Eb, thermal stress energy Et caused by heat flux, and external load stress energy Es. For example, in a typical operating environment of an automotive camera, when the local heat flux density reaches 0.5 W / ... When external vibration loads are present, the combined energy in some high-angle grain boundary regions may exceed 2.5 J / L. This process reaches the critical value for microcrack initiation. The effect of this step is that it enables a comprehensive assessment of the microscopic energy state of the pore wall surface, identifies the region most prone to failure under combined loads, and provides a precise location basis for subsequent targeted excitation.

[0100] Please continue reading. Figure 1 Based on the stress resonance perturbation wave field data, phase perturbation is recorded through a high-pass filter layer to obtain time-series data on the propagation and residence state of stress waves within the submicron structure.

[0101] In one embodiment of the present invention, the step of obtaining time-series data on the propagation and residence states of stress waves within a submicron structure by performing phase perturbation recording through a high-pass filter layer based on the stress resonance perturbation wave field data includes:

[0102] Optical path difference analysis was performed on the stress resonance disturbance wave field data to obtain the photoelastic property change path data;

[0103] Based on the photoelastic property change path data, the incident beam is deflected and separated by a high-pass filter layer to obtain phase perturbation intensity distribution data;

[0104] Based on the phase perturbation intensity distribution data, interferometric analysis is performed on the optical path difference variation within the submicron structure of the metallized hole wall to obtain stress wave propagation path data.

[0105] Based on the stress wave propagation path data, differential calculation is performed on the optical path difference phase change of the detection points at preset distances to obtain the stress wave residence location data.

[0106] Based on the stress wave residence location data and stress wave propagation path data, the phase change sequence of each detection point is sampled and analyzed at time intervals to obtain the time series data of the stress wave propagation and residence state within the submicron structure.

[0107] The following is a detailed description of the steps involved in the above embodiments:

[0108] Optical path difference analysis was performed on the stress resonance perturbation wavefield data to obtain the photoelastic property variation path data, which was achieved using a photoelastic interferometer. First, a stable laser source with a wavelength of 532 nm was split into a reference beam and a probe beam. The probe beam was focused to a diameter of 1 micrometer through a collimation system, precisely illuminating the target micro-blind buried aperture on the HDI board. When the beam passed through the aperture wall region, the refractive index of the material changed due to the stress resonance perturbation wavefield, causing the light propagation path to deflect and forming an optical path difference. Using a precisely controlled moving platform, the probe beam was scanned point-by-point along the circumference of the aperture wall in 2-micrometer steps, while a high-speed photodetector (sampling rate 100 MHz) recorded the optical path difference change at each point in real time. The collected optical path difference data and the stress resonance perturbation wavefield data were then subjected to spatiotemporal correlation analysis to establish a mapping relationship between the stress field and the optical path change, thereby obtaining the photoelastic property variation path data. For example, in an automotive camera HDI board, when the aperture wall is subjected to stress wave perturbation at a frequency of 10 MHz, every 5 MPa change in local stress will cause an optical path difference change of approximately λ / 20 (approximately 26.6 nm). This precise optical path difference analysis can non-destructively detect the dynamic stress distribution at the submicron scale, providing fundamental data for subsequent phase perturbation analysis.

[0109] Based on the photoelastic characteristic change path data, the incident beam is deflected and separated through a high-pass filter layer to obtain the phase perturbation intensity distribution data. This is accomplished using a specially designed optical filtering system. The high-pass filter layer is an optical element with spatial frequency selectivity, composed of multiple dielectric thin films, capable of blocking low spatial frequency (large-scale) light signals while allowing only high spatial frequency (small-scale) information to pass through. First, the photoelastic characteristic change path data is converted to the spatial frequency domain, and the cutoff frequency is determined to be 5 line pairs / micrometer, corresponding to sub-micrometer scale structural changes. Then, the probe beam is passed through this high-pass filter layer, achieving selective transmission of high-frequency optical path changes caused by minute stress perturbations. A phase-sensitive detector array is used to record the light intensity distribution after passing through the filter layer with a pixel pitch of 0.5 micrometers and a phase resolution of λ / 100 (approximately 5.32 nm). The phase perturbation intensity distribution data is reconstructed through inverse frequency domain transformation and phase unwrapping algorithms. In practical applications, when vehicle-mounted equipment is subjected to vibration at 50Hz and 2g acceleration, the intensity of phase perturbation generated by the metallization layer on the hole wall of an HDI board can differ by up to 10 times in different regions. Accurately identifying this difference is crucial for assessing the integrity of the hole wall. The key value of high-pass filtering technology lies in its ability to filter out large-scale background interference and highlight subtle phase changes at the submicron scale, significantly improving detection sensitivity.

[0110] Based on the phase perturbation intensity distribution data, interferometric analysis was performed on the optical path difference variation within the submicron structure of the metallized aperture wall to obtain stress wave propagation path data. This was achieved using digital holographic interferometry (DHMI). First, a two-beam interferometry system was set up to coincide the reference beam with the probe beam passing through the metallized aperture wall on a holographic recording medium, forming an interference pattern. A high-speed CCD camera (10,000 frames / second) continuously captured the dynamic changes of the interference fringes, recording the fringe movement and deformation during stress wave propagation. A time-resolved digital holographic algorithm was used to process the continuous interferometric images, extracting phase evolution information and constructing the propagation trajectory of the stress wave front. During the calculation, the phase difference between adjacent frames directly reflects the propagation speed and direction of the stress wave, while the phase gradient characterizes the intensity distribution of the stress wave. For example, in the copper metallization layer of an HDI board, the propagation speed of the stress wave is approximately 3900-4700 m / s, but significant velocity changes and reflections occur at grain boundaries; these points of change are potential stress concentration areas. This interferometric analysis method has subwavelength displacement resolution and can clearly capture the dynamic propagation process of stress waves in the frequency range of 5-100MHz, providing intuitive wave field distribution evidence for assessing the integrity of metallized hole walls.

[0111] Based on the stress wave propagation path data, the phase change of the optical path difference at detection points spaced at preset intervals is differentially calculated to obtain the stress wave residence location data. This is achieved through spatial phase analysis technology. First, multiple detection points are arranged around the metallized aperture wall, with the point spacing set to 1 / 8 to 1 / 4 of the stress wave wavelength (approximately 50-100 nanometers for a 10MHz stress wave). This spacing design ensures that spatial sampling conforms to the Nyquist sampling theorem. Then, the time series of optical path difference phase changes recorded at each detection point is processed to calculate the phase difference between adjacent detection points. φ. When a stress wave is reflected, scattered, or interfered with at a point to form a standing wave, the phase difference before and after that point will exhibit characteristic changes, such as phase jumps or amplitude amplification. By setting threshold conditions (phase gradient greater than π / 2 rad / μm or local phase change duration exceeding 5 cycles), these anomalous points, i.e., stress wave residence locations, can be identified. In actual testing of automotive HDI boards, stress wave residence mainly occurs at metal grain boundaries, metal-insulator interfaces, and regions with microscopic defects. The phase difference change amplitude at these locations is typically 3-5 times that of other regions. This differential calculation method can accurately locate the stress wave residence region, avoiding the averaging effect in traditional stress analysis, and providing an effective means to identify microstructural discontinuities.

[0112] Based on the stress wave dwell location data and stress wave propagation path data, time-series sampling analysis was performed on the phase change sequence of each detection point to obtain the time-series data of the stress wave propagation and dwell state within the submicron structure. This was accomplished using a high-temporal-resolution optical sampling system. First, based on the characteristic frequency of the stress wave (typically 5-50 MHz), the sampling rate was set to at least five times that frequency (25-250 MHz) to ensure accurate capture of waveform details. Then, time-gated optical sampling technology was used to sequentially sample each detection point at different time points during the stress wave propagation process (time intervals of 5-20 nanoseconds), recording its optical path difference phase value. Wavelet transform was applied to the acquired data for time-frequency analysis to extract the temporal evolution characteristics of the phase change. The time-series signal at the stress wave dwell location was compared with the signal in the normal propagation region to calculate the difference in spectral energy distribution and phase delay time. For example, stress wave dwell at metal grain boundaries typically manifests as energy accumulation and increased phase delay of specific frequency components, with delay times reaching 2-3 times that of the normal region. This time interval sampling analysis method can construct a complete spatiotemporal evolution map of stress waves, which not only shows where stress waves propagate and reside, but also reflects the duration of residence and the energy decay law, providing microscopic dynamic evidence for evaluating the fatigue damage accumulation mechanism of HDI boards in the vehicle environment.

[0113] Please continue reading. Figure 1 The time-series data is subjected to a two-dimensional Fourier transform to extract the spectral energy distribution parameters. The non-steady-state disturbance fingerprint is obtained by calculating the difference function through multi-point detection.

[0114] In one embodiment of the present invention, the step of performing a two-dimensional Fourier transform on the time-series data to extract spectral energy distribution parameters, and calculating the non-steady-state perturbation fingerprint through a difference function of multi-point detection, includes:

[0115] The time-series data is collected in segments according to multiple detection points along the circumferential direction of the hole wall. Combined with the grain boundary morphology characteristics of the metallized layer surface, spatial sampling sequence data related to the surface microstructure is obtained.

[0116] Based on the spatial sampling sequence data, a two-dimensional Fourier transform is performed on the time-series signal of each detection point, and the energy distribution of the metallization layer grain boundary is mapped to obtain the frequency domain main peak and grain boundary coupling distribution data.

[0117] Based on the frequency domain main peak and grain boundary coupling distribution data, the spectral tail coefficient, energy cross modulation ratio and grain boundary stress fluctuation coefficient of each detection point are extracted to obtain spectral distortion characteristic data.

[0118] By comparing the difference in spectral parameters of adjacent detection points in the spectral distortion feature data, and combining the main peak displacement, energy diffusion rate and grain boundary stress fluctuation coefficient, an unsteady-state disturbance fingerprint is obtained.

[0119] The following is a detailed description of the steps involved in the above embodiments:

[0120] The time-series data was segmented and collected from multiple probe points along the circumference of the hole wall. Combined with the grain boundary morphology characteristics of the metallized layer surface, spatial sampling sequence data related to the surface microstructure was obtained using a high-precision ring scanning system. First, 16 to 32 equally spaced probe points, with a spacing of 5-15 micrometers, were set around the metallized hole wall of the HDI board's micro-blind buried vias, covering the entire circumference of the hole wall. Then, a precision optical positioning system was used to ensure that each probe point precisely corresponded to the grain boundary morphology characteristics of the metallized layer surface. Simultaneously, a phase-shifting interferometer was used to monitor each probe point in real time, with a sampling frequency set to 200 MHz to capture microsecond-level stress fluctuations. For each probe point, time-series data for a duration of 100-500 microseconds was recorded, and the corresponding grain boundary type (e.g., high-angle grain boundary, low-angle grain boundary, or grain boundary trifle) was marked. For example, for a blind buried hole with a diameter of 80 micrometers, 24 uniformly distributed detection points are set up, with adjacent points spaced approximately 10.5 micrometers apart. Each point collects 250 microseconds of time-series data. This sampling density not only completely covers the perimeter of the hole wall but also achieves a precise correspondence with the grain boundary distribution. The effect of this segmented acquisition method is that it realizes a spatial correlation mapping between time-series data and microstructural features, providing fundamental data for subsequent analysis of the differences in stress wave response among different grain boundary types.

[0121] Based on the spatial sampling sequence data, a two-dimensional Fourier transform was performed on the time-series signal at each detection point, and the energy distribution of the metallization layer grain boundaries was mapped to obtain the frequency domain main peak and grain boundary coupling distribution data. This was accomplished using a high-performance signal processing system. First, the time-series signal at each detection point was treated as two-dimensional data (time dimension and spatial location dimension), and a two-dimensional fast Fourier transform (2D-FFT) algorithm was applied to transform the signal from the spatiotemporal domain to the frequency-wavenumber domain. In the transform, the time window was set to 200 microseconds, the spatial window covered the entire circumference of the aperture wall, and the Hanning window function was used to reduce spectral leakage effects. Then, the obtained spectrum was spatially registered with the previously acquired metallization layer grain boundary energy distribution data to establish the correspondence between frequency domain features and grain boundary structures. In the spectral analysis, the main peaks in the 1-50MHz frequency range were extracted, and their frequency position, amplitude, and full width at half maximum (FWHM) were recorded. For example, in automotive HDI board testing, high-angle grain boundary regions typically exhibit a dominant frequency response of 12-18 MHz, while grain boundary triangulation points show multi-peak characteristics of 15-25 MHz, with peak intensities often 2-3 times higher than those in ordinary regions. This two-dimensional Fourier analysis method can simultaneously reveal the distribution characteristics of stress waves in both time and space, directly correlating the stress response of micro-blind buried vias with their microstructure, providing an effective means to identify abnormal spectral characteristics caused by structural inhomogeneities.

[0122] Based on the frequency domain main peak and grain boundary coupling distribution data, the spectral tailing coefficient, energy cross-modulation ratio, and grain boundary stress fluctuation coefficient of each detection point are extracted to obtain spectral distortion characteristic data. This is achieved through a spectral feature analysis algorithm. First, the spectral tailing coefficient is calculated, which is the ratio of the energy at the tail of the main peak attenuation curve (where the frequency is higher than three times the bandwidth of the main peak) to the energy of the main peak. This parameter reflects the degree of high-frequency energy transfer caused by nonlinear effects. Then, the energy cross-modulation ratio is calculated. By analyzing the energy distribution ratio between the main frequency and its harmonics, the interaction strength between frequency components is quantified. Finally, the grain boundary stress fluctuation coefficient is calculated, and the spectral characteristics of each detection point are correlated with the grain boundary energy state at that point to establish a quantitative relationship between stress response and microstructure. In data processing, a wavelet transform enhancement algorithm is used to improve spectral resolution, especially for high-frequency components above 10MHz. For example, in the normal metallized hole wall region, the spectral tailing coefficient is usually 0.05-0.15, while in the region with microcracks, this coefficient increases to 0.3-0.5; similarly, the grain boundary stress fluctuation coefficient is 2-4 times higher in the defect region than in the normal region. These three parameters together constitute the spectral distortion characteristic data, which can comprehensively characterize the nonlinear dynamic response caused by microscopic anomalies inside the material, and provide multi-dimensional spectral evidence for accurately identifying potential failure areas of HDI boards.

[0123] The non-steady-state disturbance fingerprint is obtained by comparing the spectral parameters of adjacent detection points in the spectral distortion feature data and combining the main peak displacement, energy diffusion rate, and grain boundary stress fluctuation coefficient. This is achieved through difference analysis and pattern recognition techniques. First, the spectral parameter differences between adjacent detection points are calculated, including the main peak displacement (difference in dominant frequency between adjacent points), energy diffusion rate (rate of change in spectral width), and the gradient of the grain boundary stress fluctuation coefficient. Then, key threshold conditions are set: when the main peak displacement is greater than 1.5MHz, the energy diffusion rate exceeds 30%, or the gradient of the grain boundary stress fluctuation coefficient is greater than 0.2 / μm, it is determined to be an abnormal region. For regions that meet the conditions, points with similar parameter characteristics are classified using an adaptive clustering algorithm to form a complete outline of the non-steady-state disturbance region. In the specific implementation process, Principal Component Analysis (PCA) dimensionality reduction technology is used to map multidimensional spectral parameters to a two-dimensional feature space, generating an intuitive "non-steady-state disturbance fingerprint" map. For example, in tests where the ambient temperature in an automotive environment rapidly drops from 85°C to -20°C, 10% of the hole wall areas in the HDI board exhibit significant unsteady-state responses. These areas display a unique "fingerprint" pattern on the spectral parameter difference map, with a correspondence rate of over 90% with actual defect locations. This difference comparison method is particularly suitable for detecting local anomaly areas, eliminating interference from systematic noise and global variations, highlighting local stress anomalies caused by microstructural discontinuities, and is of significant value for predicting potential failure locations of HDI boards under automotive environments.

[0124] In one embodiment of the present invention, the step of comparing the difference in spectral parameters of adjacent detection points in the spectral distortion feature data, and combining the main peak displacement, energy diffusivity, and grain boundary stress fluctuation coefficient to obtain an unsteady-state perturbation fingerprint includes:

[0125] Optical transmittance response analysis was performed on the spectral distortion characteristic data, and stress-optical coupling characteristic data was obtained by combining the photoelectric polarization effect of the metallization layer.

[0126] Based on the stress-optical coupling characteristic data, the optical path difference variation of the micro-aperture array and the imaging sharpness attenuation coefficient are calculated, and combined with the signal transmission bandwidth requirements, photoelectric signal integrity data are obtained.

[0127] Quantum tunneling conductivity analysis is performed on regions in the photoelectric signal integrity data where the signal-to-noise ratio is lower than a second preset threshold. Combined with the interface electronic state density distribution, interface quantum transport characteristic data is obtained.

[0128] Based on the interface quantum transport characteristic data, combined with the main peak displacement, energy diffusion rate and the grain boundary stress fluctuation coefficient, an unsteady-state perturbation fingerprint is obtained.

[0129] The following is a detailed description of the steps involved in the above embodiments:

[0130] Optical transmittance response analysis was performed on the spectral distortion characteristic data, combined with the photoelectric polarization effect of the metallization layer, to obtain stress-optical coupling characteristic data. This was achieved through a precision optical transmission system and a photoelastic measurement device. First, a stable beam was generated using a monochromatic light source (a helium-neon laser with a wavelength of 633 nm), and then expanded by a beam expander to cover the entire HDI board test area. Then, after the beam passed through the test sample, a polarization optics system was used to analyze the transmitted light, recording changes in intensity and polarization state. The photoelectric polarization effect refers to the phenomenon where changes in charge distribution caused by stress affect the optical properties of a material. By measuring the changes in transmitted light characteristics under modulated electric fields at different frequencies (1-100 MHz), the photoelectric polarization response data of the metallization layer under stress was obtained. During data processing, correlation analysis was performed between the frequency components in the spectral distortion characteristic data and the frequency spectrum of transmitted light intensity changes to establish a correspondence between stress wave disturbance and optical response. For example, in the testing of HDI boards for automotive cameras, when the metallized via walls are subjected to a 15MHz stress wave disturbance, the polarization angle of the transmitted light exhibits a periodic change of 0.1-0.3 degrees. This minute change can be accurately detected using a high-precision polarization analyzer (sensitivity 0.01 degrees). This optical transmittance response analysis method overcomes the limitations of traditional electrical testing, enabling the direct capture of microstructural changes caused by stress from an optical perspective, providing a completely new dimension of analytical data for assessing the internal stress state of automotive camera HDI boards.

[0131] Based on the stress-optical coupling characteristic data, the optical path difference variation and imaging sharpness attenuation coefficient of the micro-aperture array are calculated. Combined with the signal transmission bandwidth requirements, photoelectric signal integrity data is obtained, which is achieved through an optical imaging quality analysis system. First, an optical propagation model of the micro-aperture array is constructed, treating each micro-blind hole as an optical signal transmission channel. The impact of hole wall stress fluctuations on the optical path difference is calculated based on the stress-optical coupling characteristic data. The optical path difference variation refers to the change in optical path length caused by stress in the propagation medium, directly affecting signal transmission quality. Using the Optical Transfer Function (OTF) analysis method, the modulation transfer function (MTF) curve of the micro-aperture array under stress fluctuations is calculated, and the attenuation coefficient of the MTF curve at a specific spatial frequency is extracted. This coefficient reflects the degree of decrease in imaging sharpness. Simultaneously, according to the signal transmission bandwidth requirements of the vehicle-mounted camera (typically 1.5-2.5 Gbps), optical performance indicators are converted into electrical signal quality parameters, including signal rise time, jitter, and eye diagram aperture. For example, for an 8-megapixel automotive camera, when the MTF attenuation in a certain area of ​​the HDI board exceeds 15%, the corresponding signal integrity indicators (such as eye diagram aperture) will decrease by more than 20%, approaching the boundary of the system tolerance. This calculation method links optical performance with electrical performance, establishing a complete mapping from microscopic stress disturbances to macroscopic signal quality, which has direct guiding significance for ensuring the image acquisition quality of automotive cameras in complex working environments.

[0132] Quantum tunneling conductivity analysis is performed on regions in the optoelectronic signal integrity data with a signal-to-noise ratio below a second preset threshold. Combined with the interface electronic density of states distribution, interface quantum transport characteristic data are obtained. This is achieved using a scanning tunneling electron microscope (STM) and a cryogenic electron transport measurement system. First, regions with a signal-to-noise ratio below the second preset threshold (usually set to 10 dB) are selected from the optoelectronic signal integrity data. These regions often correspond to areas of interface structural anomalies. Then, STM technology is used to scan these regions with atomic-level precision, measuring the local electronic density of states distribution at the interface to obtain quantum tunneling conductivity characteristics. Quantum tunneling conductivity refers to the non-classical transport phenomenon of electrons through a potential barrier, and its magnitude is directly related to the barrier height, width, and electronic density of states. During the measurement process, a bias voltage in the range of -0.5V to +0.5V is applied in 10mV steps, and the IV curve of the tunneling current is recorded simultaneously. The differential conductance dI / dV is calculated from the change in the slope of the curve, and this value directly reflects the interface electronic density of states distribution. For metal-insulator interfaces in automotive camera HDI boards, the quantum tunneling conductance of healthy interfaces typically exhibits a smooth bandgap structure, while interfaces with microscopic defects show discrete localized states within the bandgap, resulting in sharp peaks in the conductance curve. This quantum tunneling conductance analysis method overcomes the limitations of traditional macroscopic conductance measurements, enabling the detection of nanoscale interface defects and providing quantum mechanical evidence for the evaluation of the microstructure of metallized hole walls in submicron-level HDI boards.

[0133] Based on the interface quantum transport characteristic data, combined with the main peak displacement, energy diffusion rate, and grain boundary stress fluctuation coefficient, an unsteady-state perturbation fingerprint is obtained through multi-parameter fusion analysis and machine learning algorithms. First, a multi-dimensional feature space is established containing four key characteristic parameters: interface quantum transport characteristics (abnormal peak intensity of quantum tunneling conductance), main peak displacement (the offset of the spectral main peak relative to the standard position), energy diffusion rate (the proportion of spectral energy diffused from the main peak to the high-frequency region), and grain boundary stress fluctuation coefficient (the degree of fluctuation in stress wave amplitude at the grain boundary). Then, principal component analysis (PCA) is used to reduce the dimensionality of these four parameters, extracting the main variation directions, and constructing an "unsteady-state perturbation fingerprint" map on a two-dimensional feature plane. During pattern recognition, the support vector machine (SVM) algorithm is used to classify the fingerprint map, setting decision boundaries to distinguish samples into three categories: "stable," "metastable," and "unstable." For example, in actual testing of HDI boards for automotive cameras, when the abnormal peak value of quantum tunneling conductance exceeds three times the background value, and the main peak displacement is greater than 2 MHz with an energy diffusion rate exceeding 40%, this region is marked as "unstable," indicating a high risk of failure. This multi-parameter fusion method comprehensively considers microscopic quantum effects and macroscopic fluctuation characteristics, forming a unique "unstable perturbation fingerprint." It can accurately identify the HDI board regions most prone to failure under extreme temperature cycling and vibration environments, significantly improving the accuracy and sensitivity of automotive camera reliability prediction.

[0134] Please continue reading. Figure 1 Based on the unsteady-state perturbation fingerprint, a periodic gated pulse current is applied to the micropore region, the resistance change trajectory is collected, and charge hysteresis response data is obtained.

[0135] In one embodiment of the present invention, the step of applying a periodic gated pulse current to the micropore region based on the unsteady-state perturbation fingerprint, acquiring the resistance change trajectory, and obtaining charge hysteresis response data includes:

[0136] Based on the unsteady-state perturbation fingerprint, the stress concentration point distribution in the micropore region is extracted to obtain the electron transport channel data of the metallization layer;

[0137] The electron transport channel is partitioned and mapped according to the stress gradient direction, and a frequency-modulated gated pulse current is applied to obtain charge migration polarization intensity data.

[0138] Based on the charge migration polarization intensity data, the modulation effect of the metal grain boundary stress field on electron scattering is calculated, and the scattering modulation coefficient data is obtained.

[0139] Based on the scattering modulation coefficient data, the quantum tunneling probability and interface barrier height during electron transport are calculated to obtain quantum transport characteristic data;

[0140] Based on the quantum transport characteristic data, the coupling effect of electron scattering and quantum tunneling is calculated to obtain charge hysteresis response data.

[0141] The following is a detailed description of the steps involved in the above embodiments:

[0142] Based on the described unsteady-state perturbation fingerprint, the stress concentration point distribution in the micropore region is extracted to obtain electron transport channel data of the metallization layer. This is achieved through high-resolution stress distribution image processing and electron transport channel identification technology. First, the unsteady-state perturbation fingerprint data is converted into a stress intensity gradient map. The watershed segmentation algorithm is used to identify local maxima in the stress field; these points are the stress concentration regions. Stress concentration points are locations where the stress value is significantly higher than the surrounding area, typically occurring at metal grain boundaries, material defects, or geometric discontinuities. A stress threshold of 2.5 times the average value of the surrounding area is set to determine the boundaries of stress concentration points. Then, based on the spatial distribution of these stress concentration points, the minimum energy path algorithm is used to calculate the dominant electron transport channels in the metallization layer. Electron transport channels refer to the preferred paths for electron flow in a conductor, and their distribution is influenced by both the microstructure and the stress state. For example, in the 80-micron diameter micro-blind buried vias of an automotive camera HDI board, when the board undergoes temperature cycling from -40°C to 125°C, 3-5 major stress concentration regions typically form in the metallization layer of the via wall. These regions alter the local electron transport path, causing a 20-30% local deviation in the originally uniformly distributed current density. The effect of this step is to accurately identify the impact of micro-stress distribution on the electron transport channel, providing precise spatial positioning information for subsequent electrical testing and avoiding the blindness of traditional random sampling methods.

[0143] The electron transport channels are mapped to regions according to the stress gradient direction, and frequency-modulated gated pulse currents are applied to obtain charge migration polarization intensity data. This is achieved using a precision microelectrode array system and frequency-modulated current measurement technology. First, a microelectrode array is arranged on the surface of the HDI board with an electrode spacing of 50-100 micrometers to ensure coverage of the key areas of the identified electron transport channels. Then, the transport channels are divided into three categories according to the stress gradient direction: high-stress region, transition region, and low-stress region, corresponding to different ranges of stress gradient values ​​(high-stress region > 0.5 MPa / μm, transition region 0.1-0.5 MPa / μm, low-stress region < 0.1 MPa / μm). A precision current source is used to generate gated pulse currents modulated within the 10kHz-5MHz frequency range. These pulses feature varying duty cycles across different frequency ranges (30% for 10kHz-100kHz, 50% for 100kHz-1MHz, and 70% for 1MHz-5MHz). This selection of frequency and duty cycle takes into account the differences in electron relaxation time at metal grain boundaries under different stress states. The applied current amplitude is controlled within the 1-10mA range to avoid localized thermal effects without altering the material's intrinsic properties. The voltage response waveform is recorded in real-time using a high-speed oscilloscope (sampling rate ≥1GS / s), and the charge migration polarization intensity—the degree to which charges deviate from their equilibrium position under the influence of a local electric field—is calculated. For example, in temperature cycling tests, the charge migration polarization intensity in the high-stress region of the micro-blind buried via is 40-60% higher than that in the low-stress region, exhibiting a significant resonance peak in the 1-3MHz frequency range. This frequency-modulated electrical measurement method can distinguish subtle differences in electronic transport characteristics under different stress states, providing dynamic evidence for potential electrical failure mechanisms in automotive camera HDI boards.

[0144] Based on the charge migration polarization intensity data, the modulation effect of the metal grain boundary stress field on electron scattering is calculated, and the scattering modulation coefficient data is obtained through electron scattering theoretical analysis and numerical calculation methods. First, a local electron scattering model at the metal grain boundary is established based on the charge migration polarization intensity data. Electron scattering refers to the physical process in which free electrons change their direction of motion at lattice defects, impurities, or grain boundaries; the scattering intensity directly affects the material's conductivity. The influence of the stress field on the interaction between electrons and grain boundaries is analyzed for charge migration responses at different frequencies. A scattering modulation coefficient is introduced, defined as the ratio of the scattering rate under stress to the scattering rate under no stress; this coefficient reflects the degree of modulation of the electron scattering process by stress. The calculation of the scattering modulation coefficient considers factors such as grain boundary barrier deformation caused by stress, changes in local charge density, and enhanced phonon scattering. For example, in the copper metallization layer of an HDI board, when the local stress reaches 150 MPa, the scattering modulation coefficient at the grain boundary can increase to 1.4-1.8, meaning that the electron scattering rate increases by 40-80%, directly leading to an increase in local resistivity. For automotive camera applications, this increase in the scattering modulation coefficient results in an additional 3-5 dB signal attenuation during high-frequency signal transmission, particularly noticeable in extreme temperature environments. The benefit of this step lies in revealing how stress states alter electron scattering behavior from a microscopic physical perspective, establishing a quantitative relationship between stress, scattering, and conductivity, and overcoming the limitation of traditional macroscopic conductivity measurements that cannot distinguish contributions from different scattering mechanisms.

[0145] Based on the scattering modulation coefficient data, the quantum tunneling probability and interface barrier height during electron transport are calculated to obtain quantum transport characteristic data. This is achieved through a quantum transport analysis system and a cryogenic electron transport measurement device. First, regions where significant changes in the grain boundary barrier occur are identified based on the scattering modulation coefficient data. These regions are often key locations where electrons pass through the barrier via quantum tunneling. Quantum tunneling refers to the quantum mechanical phenomenon of electrons passing through classically forbidden regions, and its probability is closely related to the barrier height and width. A cryogenic electron transport measurement system is used to measure the nonlinear IV characteristics of the samples within a temperature range of 4.2K–300K, extracting characteristic parameters of quantum tunneling from the curve shape and temperature dependence. In data processing, the Simmons model is used to analyze the IV curves, calculating the interface barrier height (typically 0.2–0.8 eV) and the quantum tunneling probability (1–2 orders of magnitude higher in high-stress regions than in low-stress regions). For the copper-insulator interface in automotive camera HDI boards, when the stress exceeds 200 MPa, the interface barrier height decreases by 20-30%, significantly increasing the quantum tunneling probability. This change is difficult to observe at room temperature but becomes apparent under low-temperature conditions (e.g., -40°C). The effect of this step is to capture the stress-induced changes in the microscopic quantum properties of the interface, providing quantum mechanical evidence for understanding the interfacial electrical behavior at the nanoscale. This is particularly suitable for evaluating the stability of the HDI board interface microstructure under extreme temperature environments.

[0146] Based on the quantum transport characteristic data, the coupling effect of electron scattering and quantum tunneling was calculated to obtain charge hysteresis response data. This was achieved through a combination of charge dynamics analysis and a high-time-resolution electrical measurement system. First, a comprehensive charge transport model incorporating both electron scattering and quantum tunneling mechanisms was established, and their interaction under an alternating electric field was analyzed. Charge hysteresis response refers to the delayed response of charge density distribution to changes in the applied electric field; its magnitude reflects the timescale of the charge relaxation process within the material. A four-point probe resistance measurement method was used, measuring the resistance of the sample over time while applying an alternating current with a frequency of 1kHz-10MHz. The charge hysteresis response intensity was calculated by analyzing the phase difference and hysteresis loop area between the resistance-time curve and the current-time curve. Multiple sets of measurement points were set in the micro-blind buried via region, spaced 20-50 micrometers apart, forming a complete spatial distribution map of the hysteresis response. For example, in automotive HDI boards, regions where stress exceeds a critical value exhibit a significantly enhanced charge hysteresis effect, with hysteresis loop areas 2-3 times larger than normal regions, and characteristic resonance peaks appearing in the 1-3 MHz frequency range. This charge hysteresis response is particularly pronounced under the combined effects of high-frequency vehicle vibration (50-100 Hz) and alternating extreme temperatures (-40°C to 125°C), and is an important indicator for predicting the long-term reliability of HDI boards. The effect of this step is that it reveals the influence mechanism of stress state on electron transport processes from the perspective of charge dynamics, establishes a complete correlation path from microscopic quantum effects to macroscopic electrical properties, and provides unique electrical characteristic identifiers for stress failure risk assessment of automotive camera HDI boards.

[0147] Please continue reading. Figure 1 Based on the charge hysteresis response data, scanning acoustic excitation is applied to the HDI board identification area, and acoustic wave reflection intensity and phase delay data are collected to obtain acoustic pressure transition parameters.

[0148] In one embodiment of the present invention, the step of applying scanning acoustic excitation to the HDI board identification area based on the charge hysteresis response data, acquiring acoustic wave reflection intensity and phase delay data, and obtaining acoustic pressure transition parameters includes:

[0149] Based on the charge hysteresis response data, the acoustic impedance of the dislocation slip band distribution in the metallized hole wall structure of the HDI plate is calculated to obtain the acoustic impedance distribution data of micro-defects.

[0150] The scanning area is determined based on the micro-defect acoustic impedance distribution data. A scanning acoustic excitation with a frequency range of 0.5 to 3 MHz is applied. Based on the acoustic wave coupling mechanism between the metal grain boundary and the insulating medium interface, acoustic wave energy conversion data is obtained.

[0151] Based on the acoustic energy conversion data, the modulation effect of the metallization layer lattice distortion on the acoustic dispersion effect is calculated to obtain acoustic dispersion characteristic data.

[0152] The acoustic phase transition is calculated from the acoustic dispersion characteristic data, and the lattice distortion acoustoelastic response data is obtained by combining the dislocation density and acoustic wave interaction.

[0153] Based on the lattice distortion acoustic response data, the coupling parameters between microscopic acoustic wave scattering and grain boundary dislocations are extracted to obtain the acoustic pressure transition parameters.

[0154] The following is a detailed description of the steps involved in the above embodiments:

[0155] Based on the charge hysteresis response data, the acoustic impedance of the dislocation slip band distribution in the metallized hole wall structure of the HDI board was calculated to obtain the acoustic impedance distribution data of micro-defects. This was achieved through a combination of acoustic measurements and electrical data analysis. First, the charge hysteresis response data was used to identify microstructural anomaly regions, which typically contain dislocation slip bands. Dislocation slip bands refer to the banded structures formed by multiple parallel dislocation lines in a metal crystal, and are products of micro-plastic deformation. Then, a high-frequency ultrasonic probe (frequency range 20-100MHz) was used to scan these regions, with the probe diameter controlled at 30-50 micrometers to obtain sufficient spatial resolution. Based on the reflection characteristics of ultrasound in the sample, the acoustic impedance value at each point was calculated. Acoustic impedance is a physical quantity describing the resistance to sound wave propagation in a medium, equal to the product of material density and sound velocity. For normal metal grain boundaries, the acoustic impedance value changes gradually, while the dislocation slip band region exhibits abrupt changes in acoustic impedance. For example, in HDI boards for automotive cameras, after the metallized copper layer undergoes temperature cycling, the acoustic impedance value at the dislocation slip bands formed in high-stress areas decreases by 15-25% compared to the surrounding area. This difference is particularly noticeable on the walls of tiny blind vias. This acoustic impedance calculation method can accurately locate the distribution of microscopic defects in the walls of metallized vias, especially submicron-level structural anomalies that are difficult to detect in optical inspection, thus providing precise spatial positioning information for subsequent acoustic analysis.

[0156] The scanning area is determined based on the acoustic impedance distribution data of the micro-defects. A scanning acoustic excitation in the frequency range of 0.5 to 3 MHz is applied. Based on the acoustic coupling mechanism at the metal grain boundary and insulating medium interface, acoustic energy conversion data is obtained through a precision acoustic scanning system. First, based on the acoustic impedance distribution data of the micro-defects, the key area for acoustic scanning, i.e., the location of significant acoustic impedance anomalies, is determined. Then, a piezoelectric transducer generates acoustic waves in the frequency range of 0.5 to 3 MHz. This frequency range is chosen considering the matching relationship between the acoustic wavelength and the size of the micro-defects. 0.5 MHz corresponds to a wavelength of approximately 6 mm, suitable for detecting larger-scale structural changes; while 3 MHz corresponds to a wavelength of approximately 1 mm, capable of detecting finer structural anomalies. The acoustic waves are transmitted to the HDI plate through an acoustic coupling agent (usually water or a special coupling gel) and undergo reflection, refraction, and mode conversion at the metal grain boundary and insulating medium interface. A multi-channel acoustic receiver array (usually an 8×8 or 16×16 array) is used to simultaneously capture the acoustic wave response at different locations, recording the conversion of acoustic energy at different interfaces. For example, at the copper-epoxy resin interface, approximately 30-40% of the acoustic energy is reflected back to the metal layer, 50-60% is transmitted into the epoxy layer, and 5-10% is converted into interface waves. By analyzing the spatial distribution changes of these energy allocation ratios, regions with abnormal interface adhesion can be identified. This multi-frequency acoustic wave scanning technology can comprehensively evaluate the bonding quality of the metal-insulator interface in HDI boards, especially in microscopic separation regions that are prone to occur under the combined effects of vibration and temperature cycling in automotive environments.

[0157] Based on the acoustic energy conversion data, the modulation effect of metallization layer lattice distortion on acoustic dispersion was calculated, and acoustic dispersion characteristic data were obtained through an acoustic dispersion analysis system. First, the propagation speed and attenuation characteristics of acoustic waves at different frequencies were measured to establish a frequency-velocity relationship curve. Acoustic dispersion refers to the phenomenon that acoustic waves of different frequencies propagate at different speeds in a medium; this effect is particularly pronounced in materials with microstructural inhomogeneities. The phase spectral method was used, and the phase velocities of different frequency components were calculated through Fourier analysis with an accuracy of ±0.5%. Then, the measured dispersion curves were compared with the theoretical dispersion curves of ideal defect-free materials to calculate the degree of deviation and its spatial distribution. In regions containing lattice distortion, the acoustic dispersion curves exhibit characteristic distortions, such as abnormal speed reductions or changes in the slope of the dispersion curve in the 1-2 MHz frequency range. For dislocation regions in the copper metallization layer of an automotive camera HDI board, the velocity of sound waves decreases by 5-8% when propagating perpendicular to the dislocation line, while the attenuation coefficient increases by 2-3 times. This dispersion characteristic analysis method breaks through the limitation of traditional acoustic testing, which mainly relies on the intensity of sound wave reflection. By observing the changes in the dynamic characteristics of sound wave propagation in the material, it provides a more sensitive means of detecting microstructural distortions.

[0158] The acoustic wave phase transition calculation is performed on the acoustic wave dispersion characteristic data, and the lattice distortion acoustoelastic response data is obtained by combining the dislocation density and acoustic wave interaction. This is achieved through phase-sensitive acoustic analysis technology. First, time-frequency analysis is performed on the acoustic wave dispersion characteristic data to extract the phase change information of the acoustic wave during propagation. Acoustic wave phase transition refers to the phenomenon of discontinuous changes in the phase of the acoustic wave in space or time, which usually occurs at abrupt changes in material structure. The continuous wavelet transform method is used to perform fine analysis of the received acoustic wave signal with a time-frequency resolution of 1ns × 10kHz to detect phase transition events. Then, the relationship between the acoustic wave and dislocation interaction is analyzed by combining the known dislocation density distribution (usually determined by X-ray diffraction or electron backscatter diffraction). In regions with high dislocation density (e.g., > / The frequency of acoustic phase transitions is 2-4 times higher than in the normal region, and the transition amplitude (phase change) is increased by 30-50%. For example, when an automotive HDI board is subjected to a 100g acceleration vibration impact, the dislocation slip bands formed in the high-stress region of the copper metallization layer will cause an acoustic phase transition of π / 4-π / 2 radians within 10ns, while the normal region only has a transition of π / 12-π / 8 radians. This phase transition calculation method can detect minute structural changes that cannot be resolved by conventional acoustic imaging, providing a highly sensitive detection method for assessing the accumulation of microscopic damage to HDI boards under extreme operating environments.

[0159] Based on the lattice-distorted acoustoelastic response data, coupling parameters between microscopic acoustic wave scattering and grain boundary dislocations are extracted to obtain acoustic pressure transition parameters. This is achieved through acoustic parameter extraction and cluster analysis techniques. First, characteristic parameters are extracted from the lattice-distorted acoustoelastic response data, including core parameters such as phase transition rate (the number of phase transitions per unit time), transition amplitude distribution, energy attenuation coefficient, and acoustic mode conversion rate. Then, principal component analysis is used to reduce the dimensionality of these parameters, extracting the most representative acoustic feature vectors. Acoustic pressure transition parameters are comprehensive indicators characterizing the transient changes in the pressure field when sound waves propagate in a microscopically inhomogeneous medium, typically manifested as abrupt changes in sound pressure amplitude or discontinuities in sound wave phase. For the HDI board of an automotive camera, a sound pressure transition threshold is set to three times the average background value; areas exceeding this threshold are marked as high-risk areas. In practical applications, K-means clustering is used to classify all measurement points into three categories: stable region (transition parameter < 0.2), metastable region (transition parameter 0.2-0.5), and unstable region (transition parameter > 0.5). For example, after 10,000 temperature cycles, approximately 8-12% of the area at the metal-insulator interface of the micro-blind vias in the HDI board exhibits significant acoustic pressure transition characteristics, and these areas show a more than 90% agreement with subsequent actual failure locations. This acoustic pressure transition parameter analysis method bridges the gap between microscopic acoustic mechanisms and macroscopic reliability assessment, providing a quantitative criterion based on physical mechanisms for failure prediction of automotive camera HDI boards.

[0160] In one embodiment of the present invention, the step of performing nonlinear acoustoelastic analysis based on the acoustic wave dispersion characteristic data, and combining the dislocation density with the acoustic wave interaction to obtain lattice distortion acoustoelastic response data includes:

[0161] A temperature-velocity correlation mapping is established on the sound wave dispersion feature data, and the temperature-velocity coupling coefficient data is obtained by combining the sound wave propagation characteristics under extreme temperature cycles.

[0162] Based on the temperature-velocity coupling coefficient data, the acoustic phase velocity distortion of the micro-hole array under thermal stress is calculated, and the nonlinear acoustoelastic response data is obtained by combining the interface acoustic impedance jump characteristics.

[0163] The nonlinear acoustoelastic response data are subjected to acoustic wave phase-stress field-temperature field correlation analysis, and combined with the interface stress concentration factor, the interface stress-acoustic wave interaction data are obtained.

[0164] Based on the interface stress-acoustic interaction data, combined with the dislocation density distribution and lattice distortion degree, lattice distortion acoustoelastic response data are obtained.

[0165] The following is a detailed description of the steps involved in the above embodiments:

[0166] A temperature-velocity correlation mapping was established based on the acoustic dispersion characteristic data. Combined with the acoustic propagation characteristics under extreme temperature cycling, the temperature-velocity coupling coefficient data was obtained through a temperature-controlled acoustic measurement system. First, the HDI board sample was placed in a temperature-controlled chamber. A high-precision temperature controller was used to set different temperature points in 5°C steps within the range of -40°C to 125°C. After each temperature point stabilized (temperature fluctuations controlled within ±0.5°C), acoustic wave measurements were performed. Then, the ultrasonic pulse-echo technique was used to determine the propagation velocity of sound waves in the material at different temperatures. The probe frequency was selected between 1-10MHz to obtain sufficient measurement accuracy. The sound velocity measurement accuracy was controlled within ±0.1%, and the sound velocity was calculated using the time-of-flight method. For each temperature point, the sound velocity values ​​of longitudinal and transverse waves were recorded, and the gradient of sound velocity with temperature change was calculated. The temperature-velocity of sound coupling coefficient is defined as the change in relative sound velocity caused by a unit change in temperature, and is calculated using the formula: α = (1 / V) × (dV / dT), where V is the speed of sound and T is the temperature. For example, in the HDI board of an automotive camera, the temperature-velocity of sound coupling coefficient of the copper metallization layer is approximately -3 × 1 / V. / °C to -5× / °C, while the epoxy resin insulation layer is -6× / °C to -9× / °C. This coefficient difference will produce significant changes in acoustic properties at the interface between the two materials. The technical effect of this step is that it accurately quantifies the degree of influence of temperature on sound wave propagation characteristics, establishes a quantitative relationship between acoustic parameters and temperature, and provides basic data for subsequent analysis of the coupling between thermal stress and acoustic performance.

[0167] Based on the temperature-velocity coupling coefficient data, the acoustic phase velocity distortion of the micro-hole array under thermal stress is calculated. Combined with the interface acoustic impedance jump characteristics, nonlinear acoustoelastic response data is obtained through acoustoelastic analysis. First, based on the temperature-velocity coupling coefficient data and the difference in thermal expansion coefficients, the thermal stress distribution generated by the HDI board during temperature cycling is calculated. Then, the acoustoelastic principle is applied to calculate the change in acoustic phase velocity caused by stress. The acoustic phase velocity distortion is defined as the difference between the sound velocity under stress and the sound velocity in the stress-free state, and its magnitude is related to the stress level, sound wave frequency, and the acoustoelastic constant of the material. During the calculation, the anisotropic characteristics of the material are considered, and the sound velocity changes along different crystal directions are calculated separately. At the metal-insulator interface, the acoustic impedance jump characteristics are analyzed in detail, that is, the discontinuous change of acoustic impedance on both sides of the interface. In actual measurement, the interface wave technique is used to specifically detect the acoustic response of the interface region. The probe incident angle is adjusted to the critical angle for generating interface waves (usually 20°-35°), and the frequency is selected as 1-3MHz. For example, at the copper-epoxy resin interface of an HDI board, when the thermal stress reaches 150 MPa, the distortion of the acoustic phase velocity at the interface reaches 3-5% of the original sound velocity, and exhibits significant nonlinear characteristics in the high-frequency range (>2 MHz), such as a 50-80% increase in second harmonic amplitude. This nonlinear acoustoelastic response data reveals how changes in the microstructure of the material under stress affect the sound wave propagation characteristics, especially capturing minute interface changes that are difficult to detect using conventional linear acoustic methods. This has unique value for evaluating the bonding quality of the interface of automotive camera HDI boards.

[0168] The nonlinear acoustoelastic response data underwent acoustic wave phase-stress field-temperature field correlation analysis. Combined with the interface stress concentration factor, interface stress-acoustic wave interaction data was obtained using a multi-field coupling analysis system. First, the nonlinear acoustoelastic response data was spatially registered with the stress field distribution calculated by the finite element method to determine the correspondence between stress concentration regions and acoustic response anomaly regions. The interface stress concentration factor is defined as the ratio of the stress value at the interface to the far-field average stress, reflecting the degree of local stress amplification at the interface. Then, a phase-sensitive ultrasonic detection system was used to measure the relationship between the phase change of the acoustic wave during propagation and the stress field distribution. In the experiment, temperature fields with different gradients (temperature gradient range 1-5°C / mm) were applied, and the variation of the acoustic wave phase with temperature and stress was recorded. A correlation diagram of acoustic wave phase-stress-temperature was constructed using three-dimensional data visualization technology, from which the quantitative relationship among the three was extracted. For example, during the rapid cooling process from 85°C to -20°C in the automotive environment, the interfacial stress concentration factor at the copper-epoxy resin interface of the HDI board reaches 2.5-3.5, causing a change in acoustic wave phase of up to 40°-60° within 5μs, which is 4-6 times that of a uniform region. This correlation analysis method has for the first time achieved a quantitative mapping between acoustic wave response and multiple physical fields (stress field, temperature field), revealing the dynamic response characteristics of the interface microstructure under complex loads, and providing a physical basis for understanding the failure mechanism of HDI board interfaces in automotive environments.

[0169] Based on the interfacial stress-acoustic interaction data, combined with dislocation density distribution and lattice distortion degree, lattice distortion acoustoelastic response data are obtained through microscopic acoustic analysis techniques. First, the dislocation density distribution of the metallization layer is obtained through X-ray diffraction analysis, and the lattice distortion degree is calculated. Dislocation density refers to the total length of dislocation lines per unit volume, usually expressed as 1 / m²; lattice distortion degree represents the degree of deviation of the lattice from an ideal state, usually calculated through the broadening and shift of diffraction peaks. Then, spatial correlation analysis is performed between the interfacial stress-acoustic interaction data and the dislocation density distribution to determine their correspondence and correlation coefficient. During the analysis, the focus is on examining the acoustic wave response in high dislocation density regions (> / The scattering and attenuation characteristics of dislocations were studied, and the variation of scattering and attenuation coefficients with dislocation density was calculated. For the selection of the acoustic frequency, the matching relationship between the acoustic wavelength and the dislocation structure size was considered, and a frequency range of 1-3 MHz was chosen, making the wavelength approximately 0.5-2 mm, suitable for detecting micron-sized dislocation aggregates. For example, in the copper metallization layer of an automotive camera HDI board, when the dislocation density increases from... / Increase to / At this time, the acoustic attenuation coefficient increases by 3-5 times, and a significant phase abrupt change occurs along the acoustic propagation path. The magnitude of the abrupt change is positively correlated with the degree of lattice distortion. This lattice distortion acoustoel response data directly reflects the influence of microscopic dislocation structure on macroscopic acoustic performance, providing a highly sensitive detection method for the non-destructive evaluation of the microscopic damage accumulation process of HDI boards in automotive environments. This overcomes the limitation of traditional macroscopic acoustic testing, which cannot distinguish changes in microstructure.

[0170] Please continue reading. Figure 1 Based on the acoustic pressure transition parameters, the stress spectrum, hysteresis response spectrum, and acoustic pressure transition spectrum are synchronized in time and compared with energy density. The stress fluctuation risk assessment result of the micropores of the HDI plate is obtained by calculating the fluctuation spectrum coupling factor.

[0171] In one embodiment of the present invention, the step of time-synchronizing and energy-density comparing the stress spectrum, hysteresis response spectrum, and acoustic pressure transition spectrum based on the acoustic pressure transition parameters, and obtaining the stress fluctuation risk assessment result of the HDI plate micropores through fluctuation spectrum coupling factor calculation, includes:

[0172] Based on the acoustic pressure transition parameters, the stress wave resonance frequency, electron scattering frequency, and acoustic wave scattering frequency of the metallized hole wall grain boundary are phase-locked to obtain critical resonance point distribution data.

[0173] Based on the critical resonance point distribution data, the interaction strength of the dislocation stress field, charge polarization field and acoustic strain field at the metal grain boundary is calculated to obtain multi-field coupling strength data.

[0174] Based on the multi-field coupling strength data, the stress field distortion, carrier concentration gradient and acoustic dispersion coefficient at the grain boundary of the pore wall are extracted to obtain the interface bonding stability data.

[0175] Stress-temperature phase diagram analysis was performed on the micro-defects in the interface stability data, and the interface micro-failure threshold data was obtained by combining the shear slip resistance of lattice dislocations.

[0176] Based on the interface micro-failure threshold data, and combined with the cross-scale response characteristics of hole wall stress resonance, carrier migration and acoustic wave scattering, the stress fluctuation risk assessment results of HDI plate micropores are obtained.

[0177] The following is a detailed description of the steps involved in the above embodiments:

[0178] Based on the acoustic pressure transition parameters, phase locking was performed on the stress wave resonance frequency, electron scattering frequency, and acoustic wave scattering frequency of the grain boundaries of the metallized hole walls to obtain critical resonance point distribution data. This was achieved through a multi-physics synchronous measurement system. First, the micro-blind / buried via region of the HDI board was placed in a three-field joint excitation device, which consists of a piezoelectric exciter (generating 0.5-3MHz acoustic waves), a current pulse source (providing 10kHz-5MHz modulation current), and a photoelastic stress excitation system (generating stress waves). Then, the frequencies of the three physical fields were gradually changed using a frequency scanning method, with each change step being 0.05MHz, and the system response was monitored in real time. Phase locking refers to the state where the phase difference of the three physical fields remains constant, indicating that the system has achieved resonant coupling. A multi-channel phase detector was used to synchronously record the phase changes of the three fields. When the phase difference fluctuation was less than ±5° and the duration exceeded 10 cycles, phase locking was determined to have been achieved. The critical resonance point is the frequency point where the system response intensity suddenly increases, manifested as a response amplitude increase of more than 3 times and a sharp phase change (>30° / MHz). For example, in the testing of HDI boards for automotive cameras, typical critical resonance points at the interface between the copper metallization layer and epoxy resin appeared at frequencies of 1.2MHz, 2.1MHz, and 2.8MHz. These frequencies correspond to the inherent vibration modes of specific structures at the metal grain boundaries. This step, through multi-field phase-locking technology, achieved for the first time the correlation analysis of the microstructure's response to different physical fields, providing a precise means of locating potential failure regions of HDI boards from the perspective of physical field resonance.

[0179] Based on the critical resonance point distribution data, the interaction strength of the dislocation stress field, charge polarization field, and acoustic strain field at the metal grain boundaries was calculated to obtain multi-field coupling strength data, which was accomplished using a field coupling strength analysis system. First, in the identified critical resonance point region, the field strength distribution of the three physical fields was measured. The dislocation stress field was measured using a high-precision X-ray stress analyzer with a measurement accuracy of ±5 MPa; the charge polarization field was measured using a micro-area impedance analyzer to measure local resistivity changes, with a frequency range set to 10 kHz–10 MHz; and the acoustic strain field was obtained using a laser Doppler vibrometer with a vibration measurement sensitivity better than 1 nm / √Hz. Then, the inter-field coupling strength, i.e., the degree of influence of one field change on other fields, was calculated using a three-field data processing algorithm. The multi-field coupling strength is defined as the normalized response ratio caused by a unit excitation, numerically equal to the percentage change in field response divided by the percentage change in excitation. For example, in the stress concentration region of the copper-epoxy resin interface of an HDI board, when the stress field increases by 10%, the charge polarization field changes by 15-20%, the acoustic strain field changes by 12-18%, and the corresponding stress-charge coupling strength is 1.5-2.0, and the stress-acoustic coupling strength is 1.2-1.8. In the healthy region, these coupling strengths are typically less than 0.8. This multi-field coupling strength analysis method overcomes the limitations of traditional single-physical-field detection, revealing the comprehensive characteristics of material microstructure changes through the interaction between different physical fields. It is particularly suitable for evaluating the stability of HDI board interfaces under complex loads in automotive environments.

[0180] Based on the multi-field coupling intensity data, the stress field distortion, carrier concentration gradient, and acoustic dispersion coefficient at the grain boundaries of the pore walls are extracted to obtain interface bonding stability data. This is achieved through parameter extraction and interface stability assessment techniques. First, characteristic parameters of each field are separated from the multi-field coupling intensity data: the stress field distortion is defined as the deviation between the local stress tensor and the far-field average stress tensor, calculated using the principal stress ratio and directional deflection angle; the carrier concentration gradient is obtained through four-probe resistivity gradient measurement with a measurement step size of 10 μm; the acoustic dispersion coefficient is determined from the slope of wave velocity change with frequency, typically using... v / f( v represents the change in the speed of sound. f represents the frequency change. Then, a weighted summation method is used to calculate the interface bonding stability index. The weighting coefficients are determined based on the influence of each parameter on interface stability (stress field distortion 0.5, carrier concentration gradient 0.3, acoustic dispersion coefficient 0.2). The interface bonding stability index ranges from 0 to 10, with lower values ​​indicating more unstable interfaces. For example, in automotive camera HDI boards, the index for stable interfaces is typically greater than 7.5, the metastable region is 5.0-7.5, and the unstable region is less than 5.0. During temperature cycling tests from -40°C to 125°C, approximately 15-20% of the metal-insulator interface of the micro-blind buried vias in the HDI board transitions from a stable to an unstable state. This interface bonding stability analysis method comprehensively considers multiple physical characteristics, providing a quantitative evaluation of interface quality and is of significant value in predicting the long-term reliability of HDI boards in automotive environments.

[0181] Stress-temperature phase diagram analysis was performed on the microscopic defects in the interfacial bonding stability data. Combined with the shear slip resistance of lattice dislocations, the interfacial micro-failure threshold data was obtained using a thermodynamic stability analysis system. First, a fine scan was performed in the temperature-stress two-dimensional parameter space, targeting regions with an interfacial bonding stability index below 5.0. The temperature range was -40°C to 125°C, in 5°C intervals; the stress range was 0-300 MPa, in 10 MPa intervals. For each temperature-stress combination point, the change in the interfacial bonding state was measured. The stress-temperature phase diagram is a graph describing the state changes of a material under different temperature and stress conditions, similar to a traditional phase diagram, but emphasizing the influence of mechanical stress. Then, the shear slip resistance of the metal grain boundaries, i.e., the ability of grain boundaries to resist shear deformation, was determined using a nano-indenter. During testing, the indenter size was controlled between 100-500 nm to ensure that only the properties of a single grain boundary were measured. Shear slip resistance varies significantly with temperature. For example, for copper metallization layers, it is approximately 120-180 MPa at room temperature, decreases to 80-120 MPa at 125°C, and increases to 160-220 MPa at -40°C. By combining shear slip resistance data with stress-temperature phase diagrams, the critical stress threshold for interface failure at different temperatures can be determined. For instance, in the typical automotive environment temperature cycling range of -20°C to 85°C, the failure threshold for the copper-epoxy interface of an HDI board is approximately 120-160 MPa. This step provides quantitative guidance for the reliability design of HDI board interfaces by establishing the thermodynamic and mechanical conditions for interface failure.

[0182] Based on the interface micro-failure threshold data, and combined with the cross-scale response characteristics of hole wall stress resonance, carrier migration, and acoustic scattering, the stress fluctuation risk assessment results of HDI plate micropores are obtained through a cross-scale comprehensive risk assessment system. First, the interface micro-failure threshold data is compared with the load under actual working conditions to calculate the safety factor (failure threshold divided by actual working stress). Then, the cross-scale characteristics of the three physical responses are comprehensively considered: carrier migration behavior at the nanoscale, grain boundary stress resonance characteristics at the micrometer scale, and acoustic scattering modes at the millimeter scale. A cross-scale risk assessment system is established using the Analytic Hierarchy Process (AHP), and the risk indicators at each scale are weighted and fused according to their influence weights (0.3 for nanoscale, 0.5 for micrometer scale, and 0.2 for millimeter scale) to form the final risk score. The risk score ranges from 0 to 100, and the HDI plate micropore region is divided into three categories according to the score: low risk (<30), medium risk (30-70), and high risk (>70). For example, in the evaluation of a certain automotive camera HDI board, approximately 5% of 1000 micro-blind vias were rated as high-risk, 15% as medium-risk, and 80% as low-risk. High-risk areas were mainly distributed at the metal-insulator interface junctions after extreme temperature cycling (-40°C to 125°C), particularly near grain boundary junctions. This cross-scale risk assessment method is the first to achieve a full-scale mapping from microscopic physical mechanisms to macroscopic reliability, overcoming the limitations of traditional single-scale assessments and providing systematic guidance for the reliability design and process optimization of automotive camera HDI boards.

[0183] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A manufacturing process for an HDI board for an automotive camera, characterized in that, The application relates to a method for evaluating the stress wave risk of a micro-hole in an HDI plate. The method comprises the following steps: a stress resonance perturbation wave field data is obtained by perturbing the interface between a metallized hole wall and an insulating medium in the HDI plate with a short pulse light beam; time sequence data of stress wave propagation and residence in a sub-micron structure is obtained by recording the phase perturbation through a high-pass filter layer according to the stress resonance perturbation wave field data; specifically, the path data of photoelastic characteristic change is obtained by analyzing the optical path difference of the stress resonance perturbation wave field data; the phase perturbation intensity distribution data is obtained by deflecting and separating the incident light beam through the high-pass filter layer according to the path data of photoelastic characteristic change; specifically, the path data of photoelastic characteristic change is converted into a spatial frequency domain, and the cutoff frequency is 5 line pairs per micron; the detection light beam is transmitted through the high-pass filter layer to selectively transmit the high-frequency optical path change caused by the micro stress perturbation; the light intensity distribution after the high-pass filter layer is recorded, and the phase perturbation intensity distribution data is reconstructed through the frequency domain inverse transform and the phase unwrapping algorithm; wherein the high-pass filter layer is an optical element with spatial frequency selectivity, which is composed of multiple layers of dielectric film and is used to block low spatial frequency light signals and only allow high spatial frequency information to pass through; the stress wave propagation path data is obtained by interference analysis of the optical path difference change in the sub-micron structure of the metallized hole wall according to the phase perturbation intensity distribution data; the stress wave residence position data is obtained by differential calculation of the optical path difference phase change of the detection points with a preset interval; the time sequence data of stress wave propagation and residence in a sub-micron structure is obtained by time interval sampling analysis of the phase change sequence of each detection point according to the stress wave residence position data and the stress wave propagation path data; the non-steady-state perturbation fingerprint is obtained by extracting the frequency spectrum energy distribution parameter through two-dimensional Fourier transform of the time sequence data and calculating the difference function of multi-point detection; the charge hysteresis response data is obtained by collecting the resistance change trajectory by applying a periodic gate pulse current to the micro-hole region according to the non-steady-state perturbation fingerprint; the acoustic pressure transition parameter is obtained by collecting the acoustic wave reflection intensity and phase delay data by applying a scanning acoustic excitation to the identification region of the HDI plate according to the charge hysteresis response data; wherein the acoustic pressure transition parameter is a comprehensive index representing the transient change of the pressure field when the acoustic wave propagates in the micro-uniform medium, and the corresponding region is marked as a high-risk area when the acoustic pressure transition parameter exceeds the acoustic pressure transition threshold value, and the acoustic pressure transition threshold value is 3 times the average background value; the stress wave fluctuation risk evaluation result of the micro-hole in the HDI plate is obtained by time synchronization and energy density comparison of the stress spectrum, the electric hysteresis response spectrum and the acoustic pressure transition spectrum through the wave spectrum coupling factor calculation according to the acoustic pressure transition parameter; specifically, the stress wave resonance frequency, the electron scattering frequency and the acoustic wave scattering frequency of the metallized hole wall grain boundary are phase-locked to obtain the critical resonance point distribution data according to the acoustic pressure transition parameter. ​ According to the critical resonance point distribution data, the interaction intensity of the dislocation stress field, the charge polarization field and the acoustic wave strain field at the metal grain boundary is calculated, and multi-field coupling intensity data is obtained; According to the multi-field coupling intensity data, the stress field distortion variable, the carrier concentration gradient and the acoustic wave dispersion coefficient at the hole wall grain boundary are extracted, and interface bonding stability data is obtained; The stress-temperature phase diagram analysis is performed on the micro defects in the interface bonding stability data, and the shear slip resistance of the lattice dislocation is combined to obtain interface micro failure threshold data; According to the interface micro failure threshold data, combined with the cross-scale response characteristics of the hole wall stress resonance, carrier migration and acoustic scattering, the stress fluctuation risk assessment result of the HDI plate micro hole is obtained.

2. The process for manufacturing a car camera HDI board according to claim 1, wherein, The interface between the metallized hole wall and the insulating medium in the HDI plate is disturbed by a short pulse light beam, and stress resonance disturbance wave field data centered on the hole wall is obtained, including: The geometric profile of the micro blind hole in the HDI plate is profile scanned to obtain hole wall metal grain boundary distribution and orientation data; According to the metal grain boundary distribution and orientation data, the hole wall surface micro energy distribution data is obtained through grain boundary energy barrier calculation; According to the micro energy distribution data, a picosecond short pulse light beam is irradiated to the area where the energy barrier gradient exceeds the first preset threshold, and local energy excitation intensity data is obtained; According to the local energy excitation intensity data, the energy conversion coefficient of the metal grain boundary and the insulating medium bonding surface is calculated, and the interface stress expansion imbalance data is obtained; According to the interface stress expansion imbalance data combined with the energy conversion coefficient, the stress resonance disturbance wave field data centered on the hole wall is obtained.

3. The process for manufacturing a vehicle camera HDI board according to claim 2, wherein, According to the metal grain boundary distribution and orientation data, the hole wall surface micro energy distribution data is obtained through grain boundary energy barrier calculation, including: According to the metal grain boundary distribution and orientation data, temperature-stress layered analysis is established along the depth direction of the micro blind hole, and the interface stress state under extreme temperature cycling is combined to obtain grain boundary temperature stress coupling data; The temperature-stress-displacement field calculation is performed on the interface between the metallized layer and the insulating medium in the grain boundary temperature stress coupling data, and the thermal expansion coefficient gradient distribution of the metal interconnection layer is combined to obtain interface micro strain energy distribution data; According to the interface micro strain energy distribution data, the dislocation climb and slip critical energy of the metallized layer grain boundary under thermal cycle load is calculated, and the grain boundary barrier distribution map data is obtained; According to the grain boundary barrier distribution map data, combined with the heat flux density distribution and stress field intensity at the hole wall, the hole wall surface micro energy distribution data is obtained.

4. The process for manufacturing a vehicle camera HDI board according to claim 1, wherein, The two-dimensional Fourier transform is performed on the time sequence data to extract the frequency spectrum energy distribution parameter, and the non-steady-state disturbance fingerprint is obtained through the difference function calculation of multi-point detection, including: According to the spatial sampling sequence data, the two-dimensional Fourier transform is performed on the time sequence signal of each detection point, and the frequency domain main peak and grain boundary coupling distribution data are obtained by mapping the metallized layer grain boundary energy distribution; ​ According to the frequency domain main peak and the grain boundary coupling distribution data, a spectrum tailing coefficient, an energy cross modulation ratio and a grain boundary stress fluctuation coefficient of each detection point are extracted to obtain spectrum distortion characteristic data; The spectrum parameters of adjacent detection points in the spectrum distortion characteristic data are differentially compared, and the main peak displacement, the energy diffusion rate and the grain boundary stress fluctuation coefficient are combined to obtain a non-steady-state disturbance fingerprint.

5. The process for manufacturing a vehicle camera HDI board according to claim 4, wherein, The spectrum parameters of adjacent detection points in the spectrum distortion characteristic data are differentially compared, and the main peak displacement, the energy diffusion rate and the grain boundary stress fluctuation coefficient are combined to obtain a non-steady-state disturbance fingerprint, including: The spectrum distortion characteristic data is subjected to optical transmittance response analysis, and the photoelectric polarization effect of the metallization layer is combined to obtain stress-optical coupling characteristic data; According to the stress-optical coupling characteristic data, the optical path difference change of the micro-hole array and the imaging definition attenuation coefficient are calculated, and the signal transmission bandwidth requirement is combined to obtain photoelectric signal integrity data; The region with a signal-to-noise ratio lower than a second preset threshold in the photoelectric signal integrity data is subjected to quantum tunneling conductance analysis, and the interface electronic state density distribution is combined to obtain interface quantum transport characteristic data; According to the interface quantum transport characteristic data, the main peak displacement, the energy diffusion rate and the grain boundary stress fluctuation coefficient are combined to obtain a non-steady-state disturbance fingerprint.

6. The process for manufacturing a vehicle camera HDI board according to claim 1, wherein, According to the non-steady-state disturbance fingerprint, a periodic gate pulse current is applied to the micro-hole region, and a resistance change trajectory is collected to obtain charge hysteresis response data, including: According to the non-steady-state disturbance fingerprint, the stress concentration point distribution of the micro-hole region is extracted to obtain metallization layer electron transport channel data; The electron transport channel is divided and mapped according to the stress gradient direction, and a frequency-modulated gate pulse current is applied to obtain charge migration polarization intensity data; According to the charge migration polarization intensity data, the modulation effect of the metal grain boundary stress field on electron scattering is calculated to obtain scattering modulation coefficient data; According to the scattering modulation coefficient data, the quantum tunneling probability and the interface potential barrier height in the electron transport process are calculated to obtain quantum transport characteristic data; According to the quantum transport characteristic data, the coupling effect of electron scattering and quantum tunneling is calculated to obtain charge hysteresis response data.

7. The process for manufacturing a vehicle camera HDI board according to claim 1, wherein, According to the charge hysteresis response data, a scanning acoustic excitation is applied to the HDI plate recognition region, and acoustic wave reflection intensity and phase delay data are collected to obtain acoustic pressure transition parameters, including: According to the charge hysteresis response data, the dislocation slip band distribution in the HDI plate metallization hole wall structure is subjected to acoustic impedance calculation to obtain micro-defect acoustic impedance distribution data; According to the micro-defect acoustic impedance distribution data, a scanning region is determined, a scanning acoustic excitation with a frequency range of 0.5 to 3 MHz is applied, and according to the acoustic wave coupling mechanism of the metal grain boundary and the insulating medium interface, acoustic wave energy conversion data is obtained; According to the acoustic wave energy conversion data, the modulation effect of the metallization layer lattice distortion on the acoustic wave dispersion effect is calculated to obtain acoustic wave dispersion characteristic data; The acoustic wave phase transition of the acoustic wave dispersion characteristic data is calculated, and the dislocation density and the acoustic wave interaction are combined to obtain lattice distortion acoustic elastic response data; According to the lattice distortion acoustic elastic response data, a coupling parameter of micro acoustic wave scattering and grain boundary dislocation is extracted, and an acoustic pressure transition parameter is obtained.

8. The process for manufacturing a vehicle camera HDI board according to claim 7, wherein, The nonlinear acoustic elastic analysis is performed according to the acoustic wave dispersion characteristic data, the interaction of dislocation density and acoustic wave is combined, and lattice distortion acoustic elastic response data is obtained, including: A temperature-acoustic velocity correlation mapping is established for the acoustic wave dispersion characteristic data, the acoustic wave propagation characteristics under extreme temperature cycles are combined, and temperature-acoustic velocity coupling coefficient data is obtained; According to the temperature-acoustic velocity coupling coefficient data, the acoustic phase velocity distortion of the micro hole array under the action of thermal stress is calculated, and the nonlinear acoustic elastic response data is obtained in combination with the interface acoustic impedance jump characteristics; The acoustic phase-stress field-temperature field correlation analysis is performed on the nonlinear acoustic elastic response data, and the interface stress-acoustic wave interaction data is obtained in combination with the interface stress concentration coefficient; According to the interface stress-acoustic wave interaction data, the dislocation density distribution and the lattice distortion degree are combined, and the lattice distortion acoustic elastic response data is obtained.

Images