A developer for a copper foil coated with a wet film and a method

Abstract

This invention relates to the field of printed circuit board manufacturing technology, and discloses a developing and exposure machine and method for copper foil coated with a wet film. The developing and exposure machine includes a feeding and conveying mechanism, a high-frequency encoder, a broadband spectral interferometer array, a positioning vision camera, a core control unit, a DMD digital micromirror exposure engine, and a developing tank. The core control unit calculates the absolute thickness matrix and the relative solvent residue coefficient matrix based on the reflectance spectral signal to construct a nonlinear exposure compensation model; combined with the affine transformation matrix constructed by the positioning vision camera, it calculates and outputs a composite grayscale matrix. The DMD digital micromirror exposure engine controls the flipping duty cycle of the micromirror units according to the composite grayscale matrix, projecting modulated energy onto the surface of the wet film. This invention dynamically compensates for deviations in wet film thickness and solvent residue, ensuring that the crosslinking driving energy reaching the bottom layer of the copper foil is uniform and constant, and improving the verticality of the mask sidewalls.

Description

Technical Field

[0001] This invention relates to the field of printed circuit board manufacturing technology, specifically to a developing and exposure machine and method for copper foil coated with a wet film. Background Technology

[0002] In the manufacturing process of printed circuit boards and flexible electronic substrates, it is usually necessary to coat the copper foil surface with liquid photoresist (wet film) and perform exposure and development treatment to form the mask required for pattern transfer. Due to limitations in coating process and drying environment, the wet film applied to the copper foil surface often exhibits uneven physical thickness and inconsistent internal chemical solvent evaporation residues.

[0003] Existing exposure equipment typically employs a globally uniform, fixed energy distribution for illumination. This conventional exposure method fails to consider local variations in the wet film medium. When uniform exposure energy penetrates thicker areas or regions with higher solvent residue, the light energy attenuates significantly along the vertical depth direction of the medium, resulting in insufficient cross-linking driving energy to reach the copper foil bottom interface. This leads to incomplete photochemical cross-linking in the bottom layer of the medium in these areas. In subsequent development steps, the incompletely cross-linked wet film at the bottom is prone to excessive dissolution with the developer, causing the final cross-linked cured mask sidewalls to tilt inwards, failing to exhibit a vertical cross-sectional morphology and directly affecting the dimensional accuracy of fine linewidth processing.

[0004] Furthermore, flexible copper foil substrates are prone to tension stretching, lateral displacement, and angular deflection under the traction of the transmission mechanism. Existing fixed exposure equipment cannot dynamically acquire the real-time deformation state of the substrate, making it difficult to accurately align the energy compensation requirements of specific areas with the actual physical spatial coordinates of the deformed copper foil. This conflict between mechanical deformation and the fixed exposure mode can lead to distortion and misalignment of the edge contour of the photomask, further reducing the processing yield of micro-circuits. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a developing and exposure machine and method for copper foil coated with a wet film. This solves the problems of insufficient cross-linking of the underlying layer, non-perpendicular mask sidewalls, and failure to dynamically correct deformation and displacement of the flexible copper foil substrate during transmission when using fixed energy exposure equipment.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The first aspect of the present invention provides a developing and exposure machine for copper foil coated with a wet film, comprising:

[0008] A feeding and conveying mechanism is used to carry and convey copper foil coated with a wet film along a preset direction;

[0009] A high-frequency encoder is used to output transmission displacement clock pulses and establish a one-dimensional global mechanical tracking coordinate system in the system;

[0010] A broadband spectral interferometer array is used to project a continuous spectrum onto the copper foil coated with a wet film and to receive the reflected spectral signal.

[0011] A positioning vision camera is used to acquire reference point images of the surface of the copper foil coated with a wet film;

[0012] The core control unit establishes data communication connections with the high-frequency encoder, the broadband spectral interferometer array, and the alignment vision camera, and receives data, and performs data calculation and drive signal generation based on the reflection spectral signal and the reference point image.

[0013] The DMD digital micromirror exposure engine establishes a data communication connection with the core control unit to receive the drive signal and project modulated exposure energy onto the copper foil surface coated with wet film.

[0014] A developing tank is used to perform dissolution and film removal on the copper foil coated with a wet film after exposure treatment.

[0015] Preferably, the feeding and conveying mechanism includes a drive shaft, and the high-frequency encoder is coaxially connected to the drive shaft;

[0016] The core control unit receives the transmission displacement clock pulse and converts the transmission displacement clock pulse into a hardware trigger level signal;

[0017] The broadband spectral interferometer array receives the hardware trigger level signal and records discrete reflected light intensity data to generate the original reflected spectral intensity distribution tensor.

[0018] Preferably, the broadband spectral interferometer array generates the original reflectance spectral intensity distribution tensor;

[0019] The core control unit receives the original reflection spectral intensity distribution tensor and extracts the two-dimensional spatial light intensity discrete matrix.

[0020] The core control unit uses a two-dimensional Gaussian smoothing kernel to perform a convolution operation on the two-dimensional spatial light intensity discrete matrix, and outputs a smoothed spectral intensity distribution tensor.

[0021] Preferably, the core control unit obtains the smoothed spectral intensity distribution tensor, extracts a one-dimensional reflectance spectral curve sequence, performs an interference extremum point search operation on the one-dimensional reflectance spectral curve sequence, and calculates the absolute thickness matrix.

[0022] The core control unit extracts the light intensity data from the smoothed spectral intensity distribution tensor and calculates the relative solvent residue coefficient matrix.

[0023] Preferably, the core control unit obtains the absolute thickness matrix and the relative solvent residue coefficient matrix, and combines the absolute thickness matrix and the relative solvent residue coefficient matrix to construct a nonlinear exposure compensation model to generate an initial target exposure energy matrix;

[0024] The core control unit establishes a first-in-first-out (FIFO) storage structure, encapsulates the initial target exposure energy matrix into a compensation data packet and pushes it into the FIFO storage structure, thus constructing a feedforward data delay queue.

[0025] Preferably, the alignment vision camera acquires a local image containing the reference point image and transmits the local image back to the core control unit;

[0026] The core control unit extracts the two-dimensional pixel coordinates of the reference point image and constructs an affine transformation matrix;

[0027] The core control unit acquires the initial target exposure energy matrix, applies the affine transformation matrix to the initial target exposure energy matrix to perform spatial resampling and interpolation calculations, and generates a registration compensation matrix.

[0028] Preferably, the core control unit reads the circuit design file to be processed and generates a binarized target image matrix;

[0029] The core control unit acquires the registration compensation matrix, performs element-wise spatial Hadamard product operation on the binarized target image matrix and the registration compensation matrix, and outputs a composite grayscale matrix.

[0030] Preferably, the core control unit generates a composite grayscale matrix and sends the composite grayscale matrix to the DMD digital micromirror exposure engine;

[0031] The DMD digital micromirror exposure engine includes a micromirror array, which in turn includes micromirror units. The DMD digital micromirror exposure engine controls the flip duty cycle of the micromirror units within the exposure cycle using pulse width modulation based on the floating-point values ​​defined in the composite grayscale matrix, calculates the duration of the on-off duty cycle of the micromirror units, and synchronously drives the micromirror units to perform mechanical flipping.

[0032] Preferably, the developing solution in the developing tank reacts with the resist on the surface of the copper foil coated with wet film that has not undergone photochemical cross-linking reaction, thereby peeling off the wet film material in the area of ​​the composite grayscale matrix that has not been allocated exposure energy.

[0033] The crosslinking driving energy reaching the bottom interface of the copper foil coated with wet film remains a uniform constant in the global spatial dimension. The chemical stripping process proceeds in the vertical direction, and the crosslinked cured mask retained on the surface of the copper foil coated with wet film exhibits a vertical sidewall cross-sectional morphology.

[0034] A second aspect of the present invention provides a method for developing and exposing a copper foil coated with a wet film, comprising the following steps:

[0035] The broadband spectral interferometer array acquires the reflectance spectral signal, generates an original reflectance spectral intensity distribution tensor, and sends the original reflectance spectral intensity distribution tensor to the core control unit;

[0036] The core control unit receives the original reflectance spectral intensity distribution tensor, extracts the two-dimensional spatial light intensity discrete matrix, performs a convolution operation on the two-dimensional spatial light intensity discrete matrix using a two-dimensional Gaussian smoothing kernel, and outputs the smoothed spectral intensity distribution tensor.

[0037] The core control unit performs a decoupling operation on the smoothed spectral intensity distribution tensor, calculates the absolute thickness matrix and the relative solvent residue coefficient matrix;

[0038] The core control unit combines the absolute thickness matrix and the relative solvent residue coefficient matrix to construct a nonlinear exposure compensation model and generate an initial target exposure energy matrix.

[0039] The alignment vision camera acquires a local image containing the reference point image and transmits the local image back to the core control unit. The core control unit extracts the two-dimensional pixel coordinates of the reference point image and constructs an affine transformation matrix.

[0040] The core control unit applies the affine transformation matrix to the initial target exposure energy matrix to generate a registration compensation matrix.

[0041] The core control unit reads the circuit design file to be processed, generates a binarized target image matrix, performs element-wise spatial Hadamard product operation on the binarized target image matrix and the registration compensation matrix, and outputs a composite grayscale matrix.

[0042] The core control unit sends the composite grayscale matrix to the DMD digital micromirror exposure engine. The DMD digital micromirror exposure engine includes a micromirror array, which includes micromirror units. The DMD digital micromirror exposure engine controls the flipping duty cycle of the micromirror units according to the composite grayscale matrix, and projects modulated exposure energy onto the copper foil surface coated with wet film.

[0043] The solution in the developing tank dissolves and peels off the wet film in areas where no photocrosslinking reaction has occurred, forming the final conductive pattern mask.

[0044] This invention provides a developing and exposure machine and method for copper foil coated with a wet film. It has the following beneficial effects:

[0045] 1. This invention acquires reflectance spectral signals through a broadband spectral interferometer array and performs decoupling operations by the core control unit to calculate the absolute thickness matrix and the relative solvent residue coefficient matrix respectively. This design can simultaneously and independently resolve the morphology and thickness deviation of the wet film on the copper foil surface and the chemical solvent residue concentration distribution online, eliminating the data coupling interference that is easily generated when a single optical detection source processes complex film states, and providing accurate basic data support for subsequent nonlinear exposure energy modulation.

[0046] 2. This invention utilizes an absolute thickness matrix and a relative solvent residue coefficient matrix to construct a nonlinear exposure compensation model. The generated compensation parameters are then multiplied with the target image matrix using a Hadamard product to output a composite grayscale matrix, which is used to control the flip duty cycle of the micromirror units in the DMD digital micromirror exposure engine. This method can inject calculated excess compensation energy into areas with excessive wet film thickness or high solvent residue, offsetting dielectric absorption attenuation and decreased quantum efficiency. This ensures that the crosslinking driving energy reaching the substrate interface remains constant globally, resulting in a vertical sidewall cross-section on the crosslinked and cured mask after development and peeling, thus guaranteeing the processing accuracy of the underlying linewidth dimensions.

[0047] 3. This invention combines a one-dimensional global mechanical tracking coordinate system established by a high-frequency encoder with an affine transformation matrix constructed from the reference point image extracted by the alignment vision camera. Spatial resampling and interpolation calculation are performed on the initial target exposure energy matrix. This dynamic coordinate correction mechanism corrects the lateral offset, angular deflection and slight tensile deformation of the flexible copper foil substrate under the tension traction of the transmission mechanism. It ensures that the defect compensation energy of the feedforward measurement can be strictly aligned with the spatial coordinates of the actual exposure area, and prevents abnormal cross-linking of the pattern edge caused by energy compensation misalignment. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of the hardware system architecture of the developing and exposure machine for copper foil coated with a wet film according to the present invention.

[0049] Figure 2 This is a flowchart of the method of the present invention;

[0050] Figure 3 This is a sub-flowchart of the high-frequency triggering and acquisition steps of the spectral signal in this invention;

[0051] Figure 4 This is a sub-flowchart of the spatial domain low-pass filtering and noise reduction processing steps of the present invention;

[0052] Figure 5 This is a sub-flowchart of the dual-frequency domain decoupling and parameter extraction steps of the present invention;

[0053] Figure 6 This is a sub-flowchart of the nonlinear exposure energy calculation and feedforward buffering steps of the present invention;

[0054] Figure 7 This is a sub-flowchart of the spatial dynamic registration and geometric compensation steps of the present invention;

[0055] Figure 8 This is a sub-flowchart of the static graphics and dynamic parameter fusion steps of the present invention;

[0056] Figure 9 This is a sub-flowchart of the DMD exposure engine driving and pulse width modulation exposure steps of the present invention;

[0057] Figure 10 This is a sub-flowchart of the development process and undercut suppression principle of the present invention. Detailed Implementation

[0058] 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 some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] See Figure 1 The present invention provides a developing and exposure machine for copper foil coated with a wet film, comprising:

[0060] The components include a feeding and conveying mechanism, a high-frequency encoder, a broadband spectral interferometer array, a positioning vision camera, a core control unit, a DMD digital micromirror exposure engine, and a developing tank.

[0061] The feeding and conveying mechanism is used to carry and transport copper foil coated with a wet film along a preset direction. A high-frequency encoder is coaxially connected to the drive shaft of the feeding and conveying mechanism to output transmission displacement clock pulses and establish a one-dimensional global mechanical tracking coordinate system in the system.

[0062] A broadband spectral interferometer array is positioned across the feed conveyor at the feed plate end, spanning the transmission direction of the feed conveyor mechanism. It is used to project a continuous spectrum onto the copper foil coated with a wet film and to receive the reflected spectral signal.

[0063] The alignment vision camera is set in the processing area where the DMD digital micromirror exposure engine is located, and is used to acquire reference point images of the copper foil surface coated with wet film.

[0064] The core control unit establishes data communication connections with the high-frequency encoder, broadband spectral interferometer array, alignment vision camera, and DMD digital micromirror exposure engine. The core control unit is equipped with a microprocessor and field-programmable gate array hardware for performing data calculations and generating drive signals.

[0065] The developing tank is located downstream of the DMD digital micromirror exposure engine. It contains developing solution and is used to dissolve and remove the wet film from the exposed copper foil.

[0066] See Figure 2 Based on the above hardware system architecture, the present invention provides a method for developing and exposing copper foil coated with a wet film, the specific workflow of which includes the following steps:

[0067] Copper foil coated with a wet film moves into the scanning field of view of the broadband spectral interferometer array along with the feeding and conveying mechanism. Under the pulse triggering of the high-frequency encoder, the broadband spectral interferometer array synchronously acquires the reflection spectrum signal, generates the original reflection spectrum intensity distribution tensor carrying spatial coordinates, and sends the tensor to the core control unit.

[0068] The core control unit receives the original reflection spectrum intensity distribution tensor and performs a low-pass filtering algorithm in the spatial domain to filter out high-frequency scattering noise introduced by the roughness of the copper foil substrate.

[0069] The core control unit performs dual-frequency domain decoupling on the filtered spectral tensor. It extracts the phase shift of the spectral interference envelope to calculate the absolute thickness matrix and extracts the baseline absolute reflectance at a specific absorption wavelength to calculate the relative solvent residue coefficient matrix.

[0070] The core control unit combines the absolute thickness matrix and the relative solvent residue coefficient matrix, and uses the built-in photochemical attenuation equation to calculate the initial target exposure energy matrix for each spatial coordinate pixel. The core control unit stores this initial target exposure energy matrix along with the position timestamp output by the high-frequency encoder into a buffer queue.

[0071] The copper foil coated with a wet film continues to be conveyed below the DMD digital micromirror exposure engine. The alignment vision camera captures the reference alignment point on the surface of the wet-film-coated copper foil and sends the image data to the core control unit. The core control unit generates an affine transformation matrix based on the displacement of the reference alignment point in the global coordinate system.

[0072] The core control unit extracts the initial target exposure energy matrix from the buffer queue, applies an affine transformation matrix to it for spatial resampling and interpolation, and generates a registration compensation matrix that is aligned with the actual coordinates of the current exposure area.

[0073] The core control unit parses the photoplot file of the circuit to be processed and generates a binarized target image matrix. The core control unit performs element-wise Hadamard product operation on the binarized target image matrix and the registration compensation matrix, and outputs a composite grayscale matrix.

[0074] The core control unit transmits the composite grayscale matrix to the DMD digital micromirror exposure engine. Based on the floating-point values ​​defined in the composite grayscale matrix, the DMD digital micromirror exposure engine uses pulse width modulation to control the flipping duty cycle of each micromirror unit in the micromirror array during the exposure cycle, projecting modulated exposure energy onto the copper foil surface coated with a wet film.

[0075] After exposure, the copper foil coated with a wet film enters the developing tank via a feeding and conveying mechanism. The solution in the developing tank dissolves and peels off the wet film in areas where no photocrosslinking reaction has occurred, forming the final conductive pattern mask.

[0076] See Figure 3 S101, the core control unit receives the displacement pulse signal output by the high-frequency encoder as it operates with the feeding and conveying mechanism.

[0077] The high-frequency encoder can be an incremental photoelectric encoder or a rotary magnetic encoder. The internal register of the core control unit is equipped with a pulse counting module, which maps the accumulated pulse count value to the absolute displacement in the feeding and conveying direction based on the nominal pulse equivalent parameters of the high-frequency encoder and the geometric diameter of the drive roller of the feeding and conveying mechanism.

[0078] Based on this, the core control unit establishes a one-dimensional global mechanical tracking coordinate system in the system with the detection centerline of the broadband spectral interferometer array as the initial reference zero position. The vertical axis of this one-dimensional global mechanical tracking coordinate system is defined as the copper foil feeding direction, denoted as the y-axis.

[0079] S102, the core control unit receives and rectifies the displacement pulse signal, converts it into a hardware trigger level signal, and transmits it directly to the broadband spectral interferometer array via an independent data cable.

[0080] The hardware trigger level signal uses a square wave pulse signal of the transistor-to-transistor logic (TTL) standard. The broadband spectrometer interferometer array is configured to operate in an external hardware trigger mode. When the underlying drive circuit of the broadband spectrometer interferometer array detects the rising edge of the hardware trigger level signal, it immediately triggers the detector to perform single-exposure integration and analog-to-digital conversion.

[0081] This hardware-triggered communication link avoids the communication delay caused by polling calls in the software system, ensuring a tight phase-locked mapping between the spectral acquisition frequency of the broadband spectral interferometer array and the actual mechanical displacement of the copper foil.

[0082] S103, the broadband spectral interferometer array has multiple line-scan hyperspectral probe assemblies arranged along a transverse span (denoted as the x-axis) perpendicular to the y-axis. The broadband spectral interferometer array vertically projects a continuous band of detection beam onto the copper foil surface coated with a wet film.

[0083] For the specific structural implementation of the continuous broadband detection light source, those skilled in the art can use a high-power halogen lamp light source component or a broadband white LED light-emitting array. The internal light-emitting principle and the continuity characteristics of the spectral distribution are well-known technologies in the field and will not be described in detail here.

[0084] After reflection and thin-film interference, the detection beam projected onto the copper foil surface returns to the broadband spectral interferometer array. The beam splitter inside the array disperses the composite reflected beam spectrally, and a photodetector records discrete reflected light intensity data within a specific wavelength range.

[0085] S104, the broadband spectral interferometer array transmits the discrete sampled data after analog-to-digital conversion to the core control unit via a gigabit Ethernet interface or a Camera Link interface. The core control unit stores the received discrete data frames in the video memory and constructs the original reflectance spectral intensity distribution tensor based on the pixel index of the photodetector and the wavelength index calibrated by the spectrometer. .

[0086] In the aforementioned original reflectance spectral intensity distribution tensor : The discrete spatial pixel coordinates of the broadband spectral interferometer array in the transmission direction across the inlet plate are limited by the lateral resolution of the probe. This represents the discrete spatial tracking coordinates in the board-entry transmission direction generated by the pulse mapping of the high-frequency encoder; This represents the discrete sampling wavelength parameters of the broadband spectral interferometer array in the wavelength detection dimension.

[0087] The element values ​​inside this tensor represent coordinates in a specific space. Location and specific sampling wavelength The core control unit uses the absolute reflected light intensity value obtained from the detection to construct a three-dimensional data structure, thereby completing the information mapping of the copper foil surface coated with a wet film, which serves as the basic data input for subsequent frequency domain decoupling and parameter extraction.

[0088] See Figure 4In step S201, the core control unit extracts the original reflectance spectral intensity distribution tensor. The copper foil underwent surface roughening treatment in the early manufacturing process to increase adhesion, resulting in a micron-level nodular structure on its substrate. This microscopic roughness causes diffuse reflection of the detection beam projected by the broadband spectral interferometer array, leading to high-frequency reflectance abrupt noise in the original spectral data in the spatial dimension. The core control unit performs layer-by-layer traversal slicing of the original tensor along the data dimension of the wavelength parameters to obtain the two-dimensional spatial intensity discrete matrix for the current wavelength channel.

[0089] S202, the core control unit reads the pre-configured two-dimensional spatial low-pass filter parameters from its internal register. The core control unit uses a two-dimensional Gaussian smoothing kernel to construct a filter matrix to achieve the two-dimensional spatial low-pass filtering function. The two-dimensional Gaussian smoothing kernel allocates the weights of neighboring pixels according to the two-dimensional normal distribution law. Its horizontal and vertical discrete distribution degree and spatial smoothing strength are limited by the preset standard deviation parameter of the filter.

[0090] S203, the core control unit uses the constructed two-dimensional Gaussian smoothing kernel to perform convolution operations on the discrete two-dimensional spatial light intensity matrices of each wavelength extracted from the slices one by one. The core control unit guides the smoothing kernel matrix to slide along the global tracking coordinates in the spatial dimension, performs a weighted average of the reflected light intensity values ​​of the target pixel and its neighboring pixels, and then filters out the spatial high-frequency abrupt signals caused by the morphology of the basal nodules, outputting the smoothed spectral intensity distribution tensor.

[0091] For the matrix edge pixel data filling and boundary extension processing involved in discrete two-dimensional convolution operations, those skilled in the art can use conventional image processing algorithms such as zero-fill edge or mirror copy edge. The basic matrix operation logic is a well-known technology in this field and will not be elaborated here.

[0092] After the core control unit completes the low-pass smoothing process in the spatial domain, the reflectivity baseline drift amplitude in the smoothed spectral intensity distribution tensor is reduced, providing a data foundation with a signal-to-noise ratio that meets the requirements for subsequent frequency domain interference fringe decoupling.

[0093] See Figure 5 S301, the core control unit reads the smoothed spectral intensity distribution tensor after spatial domain noise reduction processing. For each discrete spatial pixel coordinate in the global spatial tracking coordinate system, the core control unit performs dimensionality reduction extraction along the data dimension of the wavelength parameter to obtain the one-dimensional reflectance spectrum curve sequence at the corresponding pixel position.

[0094] S302, the core control unit performs an interference extremum point search operation on the extracted one-dimensional reflectance spectrum curve sequence.

[0095] The core control unit employs a first-order differential zero-crossing detection algorithm to perform calculations on the discrete curve sequence to search for interference extrema and locate the peaks and troughs of the interference fringes. For the zero-crossing of the first derivative of continuous signal extrema and the rules of differential operation for discrete sequences, those skilled in the art can implement these using conventional signal processing mathematical toolkit functions. The basic algorithm code structure is well-known in the field and will not be elaborated upon here.

[0096] S303, the core control unit calculates the absolute thickness of the wet film at a specific coordinate point using a thin-film interference phase model based on the obtained wavelength parameters of the interference extrema, and generates an absolute thickness matrix by traversing all spatial pixel coordinate combinations. The calculation logic for the absolute thickness of the wet film is expressed by the following formula:

[0097] ;

[0098] in, Represents pixel coordinates in discrete space The calculated absolute thickness of the wet film; This represents the interference order difference constant between two adjacent interference extrema. This indicates the wavelength parameter corresponding to the first interference extremum point located by the search; This represents the wavelength parameter corresponding to the second interference extremum point adjacent to the first interference extremum point; This represents the reference refractive index constant of the wet film photoresist coated on the copper foil surface under standard curing conditions. The core control unit obtains this parameter by reading the pre-stored material formulation file. It represents the absolute value of the difference between two adjacent wavelength parameters.

[0099] S304, while performing the aforementioned decoupling of the high-frequency characteristics of the interference phase, the core control unit performs decoupling extraction of the baseline low-frequency attenuation characteristics based on the same smooth spectral intensity distribution tensor. The core control unit locates the specific absorption wavelength characterizing the absorption properties of the coating solvent in the system formulation file, extracts the smooth light intensity data corresponding to the specific absorption wavelength from the tensor, and performs a ratio conversion calculation with the light intensity value of the reference white board obtained in the offline calibration stage to generate the absolute reflectance of the current pixel.

[0100] S305, the core control unit assesses the spectral baseline absorption loss caused by incomplete solvent evaporation based on the acquired absolute reflectance, and calculates and outputs a relative solvent residue coefficient matrix. The core control unit, combined with the linear conversion constant between reflectance attenuation and solvent concentration pre-acquired through offline system testing and calibration, maps the loss deviation between the actual absolute reflectance of the substrate extracted at a specific absorption wavelength and the reference absolute reflectance at the same wavelength under ideal, completely dry baseline conditions, determining the relative solvent residue coefficient at each discrete spatial pixel coordinate point.

[0101] By performing the aforementioned multidimensional calculation steps, the core control unit deconstructs a single data source containing broadband optical information into separate thickness parameters and solvent chemical parameters, providing data support for subsequent nonlinear exposure energy modulation.

[0102] See Figure 6 S401, the core control unit obtains the absolute thickness matrix and relative solvent residue coefficient matrix generated by the aforementioned calculations. The core control unit accesses the system's internal memory, retrieves the system formula file corresponding to the current batch of copper foil to be processed, and reads the preset photochemical basic calibration parameters.

[0103] The basic calibration parameters for photochemistry include standard wet film thickness, standard solvent residual coefficient, reference exposure energy threshold, material absorption coefficient corresponding to a specific wavelength, and quantum efficiency interference constant.

[0104] S402, the core control unit, constructs a nonlinear exposure compensation model that reflects the actual state of the medium for each discrete spatial pixel coordinate point within the global spatial tracking coordinate system. For the construction of the mathematical model for the exponential decay of the penetration depth of ultraviolet beams in media such as photoresists, those skilled in the art can refer to the classic derivation of the Beer-Lambert law. The derivation of the optical absorption cross-section distribution and the medium thickness is a well-known technique in this field and will not be elaborated upon here.

[0105] The core control unit combines the basic model of Beer-Lambert's law with the linear constraint relationship between solvent residue and photoinitiator reaction kinetics to calculate the theoretical energy required at each coordinate point.

[0106] S403, the core control unit traverses and performs calculations to generate the initial target exposure energy matrix.

[0107] The specific calculation logic for the energy values ​​of discrete pixels within this matrix is ​​expressed by the following formula:

[0108] ;

[0109] in, Represents pixel coordinates in discrete space The initial target exposure energy value calculated at the location; This indicates the baseline exposure energy threshold under standard wet film conditions as specified in the formulation document; Represents an exponential function with the natural constant as its base; This indicates the material absorption coefficient of the current wet film batch for a specific wavelength of the exposure light source; This represents the absolute thickness of the wet film obtained in the upstream data extraction step; This represents the target standard wet film thickness constant specified in the formulation document; This represents the linear adjustment constant for the interference of solvent residue on the quantum efficiency of photoinitiated dose, as specified in the formulation file. This represents the relative solvent residue coefficient obtained in the upstream data extraction step; This indicates the target standard solvent residual coefficient set in the formulation document.

[0110] S404, the core control unit has an independent data buffer configured in the internal volatile memory to establish a first-in-first-out storage structure. Considering that the displacement of the copper foil coated with wet film on the feeding and conveying mechanism takes time, the compensation energy data calculated above cannot be sent to the execution end immediately.

[0111] The core control unit encapsulates the calculated initial target exposure energy matrix, along with the high-frequency encoder displacement coordinate data corresponding to the current matrix and the system's underlying absolute timestamp, into an independent compensation data packet. The core control unit pushes this compensation data packet into the aforementioned first-in-first-out storage structure to construct a feedforward data delay queue, awaiting subsequent call requests from the downstream dynamic registration process.

[0112] See Figure 7 S501, the copper foil coated with a wet film runs with the feeding and conveying mechanism.

[0113] The core control unit receives the pulse count value output in real time from the high-frequency encoder to determine that a specific section of the copper foil has entered the processing field of view of the DMD digital micromirror exposure engine. Based on the position determination signal, the core control unit sends an image acquisition command to the alignment vision camera. The alignment vision camera acquires a local image containing the copper foil reference alignment point and sends the image data back to the core control unit.

[0114] The reference alignment point is comprised of a cross-shaped optical marker or a mechanically drilled hole etched during the previous process of the printed circuit board. The core control unit performs image parsing to extract the two-dimensional pixel coordinates of the reference alignment point in the coordinate system of the current exposure area. For the basic logic of image smoothing and denoising, edge operator extraction, and geometric centroid localization in machine vision, those skilled in the art can implement it using standard function libraries from open-source vision libraries. The basic algorithm code structure is well-known in the field and will not be elaborated upon here.

[0115] S502, due to the lateral displacement, angular deflection and slight longitudinal stretching of the flexible substrate material under the tension of the transmission mechanism, the coordinates of the previously measured data will drift.

[0116] The core control unit extracts the deviation between the current actual reference coordinates and the theoretical reference coordinates set in the front measurement area. Based on the displacement deviation parameters, the core control unit calculates the spatial geometric relationship from the coordinate system of the front measurement area to the actual coordinate system of the current exposure area, and constructs a third-order affine transformation matrix containing coordinate translation vectors, rotation deflection angles, and multi-axis scale scaling factors to complete the mapping rule transformation from the horizontal and vertical tracking coordinates of the front measurement area to the actual coordinate system of the current exposure area.

[0117] S503: The core control unit obtains the current absolute displacement coordinate parameters and uses them as a search key to access the feedforward data delay queue in the internal memory. The core control unit performs queue matching and popping operations, retrieving the corresponding initial target exposure energy matrix from the first-in-first-out buffer structure.

[0118] S504, the core control unit applies the affine transformation matrix obtained above to the initial target exposure energy matrix. The core control unit calls the internal logic gate array to perform spatial resampling and interpolation calculations, and remaps the discrete energy values ​​after geometric deformation to the regular exposure area pixel grid.

[0119] In the spatial resampling interpolation calculation, the core control unit adopts a bilinear interpolation algorithm, which takes the data of four adjacent integer pixels around the target mapping point and performs distance-weighted averaging to generate a registration compensation matrix that is closely aligned with the real space of the exposure area.

[0120] Through the aforementioned dynamic coordinate correction mechanism, the system corrects the spatial misalignment of the feedforward data caused by substrate deformation.

[0121] See Figure 8 S601, the core control unit reads the externally imported circuit design file to be processed.

[0122] The circuit design file uses industry-standard photoplotting data files in Gerber RS-274X or ODB++ format. The core control unit calls the internally configured raster image processing module to perform discretization analysis on the above vector format design file, generating a binarized target graphic matrix that matches the current actual coordinate system dimensions of the exposure area.

[0123] For polygon closure filling algorithms and vector line segment rasterization interpolation algorithms in raster image processing, those skilled in the art can implement them using conventional open-source graphics processing algorithm libraries. The underlying matrix discretization mapping logic is a well-known technology in this field and will not be elaborated here.

[0124] In the generated binary target graphic matrix, the core control unit assigns a scalar value of 1 to the pixel elements of the solid conductive graphic regions that need to undergo photochemical cross-linking reactions, and assigns a scalar value of 0 to the pixel elements of the substrate blank regions that do not need to undergo reactions.

[0125] S602, the core control unit obtains the registration compensation matrix generated by the aforementioned process and reads the maximum output energy threshold parameter from the hardware configuration file inside the system.

[0126] The specific value of the maximum output energy threshold is determined by the absolute peak light power of the ultraviolet light source built into the DMD digital micromirror exposure engine and the single control clock cycle limit set by the underlying system. The core control unit performs a data boundary threshold check operation, limits the discrete energy values ​​within the registration compensation matrix, and forcibly constrains all compensation energy values ​​to not exceed the maximum output energy threshold to avoid data overflow anomalies in subsequent underlying drive commands.

[0127] S603, the core control unit performs element-wise Hadamard product operation on the binarized target graphic matrix and the registration compensation matrix after clipping and normalization. Utilizing the mathematical property of element-wise multiplication of corresponding matrix positions, the core control unit ensures that the product result is always zero in regions where the binarized target graphic matrix is ​​assigned a value of 0, thus maintaining the two-dimensional geometric boundaries and insulation spacing of the original circuit pattern unaffected by the dynamic energy modulation process. In regions where a value of 1 is assigned, the normalized compensation energy value for the corresponding spatial coordinate point is directly loaded. The core control unit generates a composite grayscale matrix containing multi-level floating-point values ​​through the above operation, and this operational relationship is expressed by the following formula:

[0128] ;

[0129] in, This represents a composite grayscale matrix generated in the actual coordinate system of the current exposure area, whose internal elements are floating-point numbers between 0 and 1; This represents the binarized target graphic matrix generated by rasterizing the circuit design file to be processed; The Hadamard product operator represents the element-wise multiplication of two matrices of the same order. This represents the registration compensation matrix after spatiotemporal dynamic registration and affine transformation resampling processing in the aforementioned process; This represents the maximum energy threshold constant that the DMD digital micromirror exposure engine can output within a single exposure cycle.

[0130] The core control unit completes the data flow fusion of the static target image and dynamic parameters, and the generated composite grayscale matrix has both circuit geometric boundary constraints and exposure energy spatial nonlinearity compensation information, serving as the direct data input source for subsequent digital micromirror array pulse width modulation.

[0131] See Figure 9 The S701 core control unit sends the calculated composite grayscale matrix to the DMD digital micromirror exposure engine via a fiber optic data link or a high-speed serial computer expansion bus (PCIe) interface.

[0132] The DMD (Digital Micromirror Device) exposure engine is internally equipped with a high-speed field-programmable gate array (FPGA) driver board and a digital micromirror chip composed of a massive array of micron-sized reflective mirrors. The driver board receives the composite grayscale matrix containing floating-point values ​​and parses the numerical dimension information corresponding to each spatially discrete pixel in the matrix to establish a spatial pixel mapping table.

[0133] The S702 driver board is based on the pulse width modulation (PWM) working principle. It converts the obtained spatial grayscale floating-point values ​​into the underlying hardware control timing in the time domain. For the electromechanical execution principle of the micromirror unit inside the digital micromirror chip, which relies on the electrostatic force of the bottom electrode to achieve a fixed deflection angle (e.g., ±12 degrees) mechanical flip, those skilled in the art can refer to the standard driver architecture of microelectromechanical systems (MEMS). Its underlying deflection and latching characteristics are well known in the field and will not be described in detail here.

[0134] The driver board utilizes a pulse width modulation mechanism to linearly adjust the energy intensity of the spatially localized projection beam by controlling the flip-dwell time ratio of each micromirror unit within a single fixed exposure refresh cycle.

[0135] The S703 driver board calculates the on-duty cycle duration of each micromirror unit on the digital micromirror chip within the current exposure control cycle. When the micromirror unit is in the on-off deflection state, the ultraviolet beam emitted by the system's built-in light source is reflected and projected onto the copper foil surface coated with a wet film. When the micromirror unit is in the off-off deflection state, the ultraviolet beam is deflected and guided to the light-absorbing baffle inside the optical engine.

[0136] The driver board calculates the proportional mapping relationship based on the floating-point element values ​​within a specific closed interval of the composite grayscale matrix input from the core control unit, combined with the total absolute duration constant of the single exposure refresh control cycle set at the system's underlying level. It then outputs the absolute duration parameter of the micromirror unit remaining in the on state at the actual discrete coordinates of the current exposure area. .

[0137] The S704, a continuous beam of light with constant power is generated by an ultraviolet light-emitting diode array or ultraviolet laser inside the DMD digital micromirror exposure engine and uniformly projected onto the surface of the digital micromirror chip through an illumination optical path. The digital micromirror chip receives the light from the driver board according to the above... The control commands issued by the array synchronously drive millions of micromirror units to perform independent high-frequency mechanical flipping actions.

[0138] The reflected ultraviolet beam, modulated by the time-domain duty cycle, passes through the projection optical objective lens group and is projected onto the surface of the copper foil coated with a wet film, which moves continuously with the feeding and conveying mechanism. The system relies on this hardware to execute logic, converting the feedforward-sensed multidimensional defect parameters into actual photochemical energy compensation operations. While maintaining the clear two-dimensional geometric boundaries of the original circuit pattern to be processed, the system injects accurately calculated compensation exposure energy into local areas where there is thickness variance or solvent residue variance.

[0139] See Figure 10 S801, the copper foil coated with a wet film after being processed by the DMD digital micromirror exposure engine continues to be transported downstream by the feeding and conveying mechanism and enters the developing tank.

[0140] The developing tank is internally equipped with a spray array and drive rollers for spraying a weakly alkaline aqueous solution. For the spray pressure control, developer temperature regulation, and mechanical transmission structure of the drive rollers within the developing equipment, those skilled in the art can refer to the conventional parameters of standard printed circuit board wet process manufacturing equipment for setting these parameters. The fluid dynamics and mechanical transmission logic are well-known technologies in the field and will not be elaborated upon here.

[0141] The developing solution in the developing tank comes into contact with the copper foil surface coated with wet film, and reacts with the resist that has not undergone photochemical cross-linking to dissolve it, thereby peeling off the wet film material in the area of ​​the composite grayscale matrix that has not been allocated exposure energy.

[0142] S802, after being exposed to ultraviolet light, generates free radicals in its internal photoinitiator, which then initiates a polymerization and crosslinking reaction.

[0143] The spatial crosslinking density of this polymerization reaction is positively correlated with the locally absorbed light energy. Under normal conditions without spatial dynamic nonlinear compensation, if there is a positive variance in the thickness of the wet film or a high concentration of residual solvent, the actual photochemical energy reaching the copper foil substrate interface will be lower than the curing threshold due to the attenuation characteristics of light in the medium, resulting in insufficient bottom crosslinking density. This gradient crosslinking density distribution, under the lateral dissolution effect of the developer, will form an undercut morphology, leading to deviations in the final retained mask linewidth.

[0144] S803, the system executes the aforementioned spatial dynamic nonlinear compensation mechanism. The core control unit generates a composite grayscale matrix that integrates defect state parameters and drives the micromirror array. In coordinate regions where the wet film thickness is too large or the solvent residual concentration is too high, the system increases the micromirror activation duty cycle at that local location and injects excess exposure energy.

[0145] This excess exposure energy compensates for the dielectric absorption attenuation loss caused by the increased thickness and counteracts the interference of decreased quantum efficiency caused by solvent residue. The mathematical derivation logic of the system controlling the exposure energy reaching the bottom interface is expressed by the following formula:

[0146] ;

[0147] in, This represents the actual discrete spatial pixel coordinates reached after attenuation by the wet film medium. The exposure energy value corresponding to the bottom substrate interface; This indicates that the system calculates and actually injects the exposure energy value into this coordinate point based on the aforementioned nonlinear initial energy compensation matrix calculation model; Represents an exponential function with the natural constant as its base; Indicates the material absorption coefficient; This represents the absolute thickness value at the current coordinate point; The linear adjustment constant representing the interference of solvent residue on quantum efficiency; This represents the relative solvent residue coefficient at the current coordinate point; This represents the residual coefficient of the target standard solvent.

[0148] S804, substituting the aforementioned calculation formula for the initial target exposure energy matrix into the above formula, and simplifying by canceling out the variables, it can be concluded that the exposure energy reaching the bottom interface is always equal to... This indicates that, under the action of the dynamic compensation mechanism, regardless of the local variance in the three-dimensional morphology of the surface wet film and the distribution of chemical solvents, the cross-linking driving energy reaching the bottom interface of the copper foil remains a uniform constant in the global spatial dimension.

[0149] S805, when the developer dissolves the boundary of the unexposed area, the chemical stripping process proceeds uniformly in the vertical direction because the crosslinking gradient of the exposed area in the vertical direction is under control and the solvent erosion resistance at the substrate interface reaches a consistent design threshold.

[0150] After the dissolution reaction is completed, the cross-linked cured mask remaining on the copper foil surface exhibits a vertical sidewall cross-sectional morphology. Its bottom line width dimension is strictly consistent with the boundary of the binarized target pattern defined in the circuit design file to be processed, thus eliminating the influence of variance interference introduced by the pre-coating process on the development accuracy.

[0151] S901, the core control unit executes an offline optical reference calibration program when the system is not in production or processing mode. The core control unit controls the broadband spectral interferometer array to perform a scanning operation on a standard pure reflective substrate.

[0152] The system uses a polished bare copper plate or a standard white plate with a known reflectivity curve as a pure reflective substrate. The core control unit extracts the reflected light intensity signal at a specific absorption wavelength to generate a reference absolute reflectivity.

[0153] Meanwhile, operators conducted offline tests on the reflectivity attenuation of substrates coated with different concentrations of solvent, and calculated the linear conversion constant between the degree of reflectivity attenuation and solvent concentration. The core control unit then entered the above optical parameters and the reference refractive index constant of a specific wet-film photoresist under standard curing conditions into the system's underlying database.

[0154] S902, the core control unit executes an offline photochemical parameter calibration procedure to obtain various constants used to construct a nonlinear exposure compensation model. For the step exposure test and development linewidth measurement of the exposure tolerance of photoresist, those skilled in the art can use standard grayscale wedge film for offline evaluation. The basic development test and linewidth measurement methods are well-known technologies in the field and will not be described in detail here.

[0155] Based on the stepped exposure test data, the system extracts the reference exposure energy threshold, target standard wet film thickness constant, target standard solvent residue coefficient, material absorption coefficient at a specific wavelength, and linear adjustment constant of solvent residue on quantum efficiency under standard wet film conditions.

[0156] S903, the core control unit encapsulates the aforementioned optical reference calibration parameters and photochemical parameters to construct a system formulation file for a specific batch of materials. As the lower-level data structure feature of the system formulation file, it adopts a static text file in Extensible Markup Language (XML) format or JavaScript Object Notation (JSON) format and is stored in the non-volatile storage hard disk of the core control unit.

[0157] The core control unit integrates the reference refractive index constant, linear conversion constant, reference absolute reflectance value, reference exposure energy threshold, target standard wet film thickness constant, target standard solvent residual coefficient, material absorption coefficient and linear adjustment constant obtained from the previous calibration to construct a system formula parameter vector for a single type of copper foil coated with wet film, and converts the parameter vector into a column vector form for fixed storage through transpose operation.

[0158] S904, during the production start-up phase, the core control unit receives the current production batch number of copper foil to be processed from the factory manufacturing execution system (MES). The core control unit uses this production batch number as the index primary key to search and match the corresponding system recipe file in the underlying database.

[0159] The core control unit parses the parameter vector within the file and loads it from the non-volatile hard disk into static random access memory (SRAM). During subsequent continuous transmission and processing, the frequency domain decoupling extraction module of the broadband spectral interferometer array and the energy compensation matrix calculation module of the DMD digital micromirror exposure engine directly read the corresponding constant values ​​from this SRAM for real-time calculation. This recipe calling logic ensures that when switching between batches of copper foil with different thicknesses and material properties, the system's underlying calculation model does not need to be recompiled, maintaining the continuity of the data processing chain.

Claims

1. A developing and exposure machine for copper foil coated with a wet film, characterized in that, include: A feeding and conveying mechanism is used to carry and convey copper foil coated with a wet film along a preset direction; A high-frequency encoder is used to output transmission displacement clock pulses and establish a one-dimensional global mechanical tracking coordinate system in the system; A broadband spectral interferometer array is used to project a continuous spectrum onto the copper foil coated with a wet film and to receive the reflected spectral signal. A positioning vision camera is used to acquire reference point images of the surface of the copper foil coated with a wet film; The core control unit establishes data communication connections with the high-frequency encoder, the broadband spectral interferometer array, and the alignment vision camera, and receives data, and performs data calculation and drive signal generation based on the reflection spectral signal and the reference point image. The DMD digital micromirror exposure engine establishes a data communication connection with the core control unit to receive the drive signal and project modulated exposure energy onto the copper foil surface coated with wet film. A developing tank is used to perform dissolution and film removal on the copper foil coated with a wet film after exposure treatment.

2. The developing and exposure machine for copper foil coated with a wet film according to claim 1, characterized in that, The feeding and conveying mechanism includes a drive shaft, and the high-frequency encoder is coaxially connected to the drive shaft; The core control unit receives the transmission displacement clock pulse and converts the transmission displacement clock pulse into a hardware trigger level signal; The broadband spectral interferometer array receives the hardware trigger level signal and records discrete reflected light intensity data to generate the original reflected spectral intensity distribution tensor.

3. The developing and exposure machine for copper foil coated with a wet film according to claim 2, characterized in that, The broadband spectral interferometer array generates the original reflection spectral intensity distribution tensor; The core control unit receives the original reflection spectral intensity distribution tensor and extracts the two-dimensional spatial light intensity discrete matrix. The core control unit uses a two-dimensional Gaussian smoothing kernel to perform a convolution operation on the two-dimensional spatial light intensity discrete matrix, and outputs a smoothed spectral intensity distribution tensor.

4. A developing and exposure machine for copper foil coated with a wet film according to claim 3, characterized in that, The core control unit obtains the smoothed spectral intensity distribution tensor, extracts the one-dimensional reflectance spectral curve sequence, performs an interference extreme point search operation on the one-dimensional reflectance spectral curve sequence, and calculates the absolute thickness matrix. The core control unit extracts the light intensity data from the smoothed spectral intensity distribution tensor and calculates the relative solvent residue coefficient matrix.

5. A developing and exposure machine for copper foil coated with a wet film according to claim 4, characterized in that, The core control unit obtains the absolute thickness matrix and the relative solvent residue coefficient matrix, and combines the absolute thickness matrix and the relative solvent residue matrix to construct a nonlinear exposure compensation model to generate the initial target exposure energy matrix; The core control unit establishes a first-in-first-out (FIFO) storage structure, encapsulates the initial target exposure energy matrix into a compensation data packet and pushes it into the FIFO storage structure, thus constructing a feedforward data delay queue.

6. A developing and exposure machine for copper foil coated with a wet film according to claim 1, characterized in that, The alignment vision camera acquires a local image containing the reference point image and transmits the local image back to the core control unit; The core control unit extracts the two-dimensional pixel coordinates of the reference point image and constructs an affine transformation matrix; The core control unit acquires the initial target exposure energy matrix, applies the affine transformation matrix to the initial target exposure energy matrix to perform spatial resampling and interpolation calculations, and generates a registration compensation matrix.

7. A developing and exposure machine for copper foil coated with a wet film according to claim 6, characterized in that, The core control unit reads the circuit design file to be processed and generates a binary target image matrix. The core control unit acquires the registration compensation matrix, performs element-wise spatial Hadamard product operation on the binarized target image matrix and the registration compensation matrix, and outputs a composite grayscale matrix.

8. A developing and exposure machine for copper foil coated with a wet film according to claim 7, characterized in that, The core control unit generates a composite grayscale matrix and sends the composite grayscale matrix as the driving signal to the DMD digital micromirror exposure engine. The DMD digital micromirror exposure engine includes a micromirror array, which in turn includes micromirror units. The DMD digital micromirror exposure engine controls the flip duty cycle of the micromirror units within the exposure cycle using pulse width modulation based on the floating-point values ​​defined in the composite grayscale matrix, calculates the duration of the on-off duty cycle of the micromirror units, and synchronously drives the micromirror units to perform mechanical flipping.

9. A developing and exposure machine for copper foil coated with a wet film according to claim 8, characterized in that, The developing solution in the developing tank reacts with the resist on the surface of the copper foil coated with wet film that has not undergone photochemical cross-linking, thus peeling off the wet film material in the area of ​​the composite grayscale matrix that has not been allocated exposure energy.

10. A method for developing and exposing copper foil coated with a wet film, applied to a developing and exposing machine for copper foil coated with a wet film according to any one of claims 1-9, characterized in that, Includes the following steps: The broadband spectral interferometer array acquires the reflectance spectral signal, generates an original reflectance spectral intensity distribution tensor, and sends the original reflectance spectral intensity distribution tensor to the core control unit; The core control unit receives the original reflectance spectral intensity distribution tensor, extracts the two-dimensional spatial light intensity discrete matrix, performs a convolution operation on the two-dimensional spatial light intensity discrete matrix using a two-dimensional Gaussian smoothing kernel, and outputs the smoothed spectral intensity distribution tensor. The core control unit performs a decoupling operation on the smoothed spectral intensity distribution tensor, calculates the absolute thickness matrix and the relative solvent residue coefficient matrix; The core control unit combines the absolute thickness matrix and the relative solvent residue coefficient matrix to construct a nonlinear exposure compensation model and generate an initial target exposure energy matrix. The alignment vision camera acquires a local image containing the reference point image and transmits the local image back to the core control unit. The core control unit extracts the two-dimensional pixel coordinates of the reference point image and constructs an affine transformation matrix. The core control unit applies the affine transformation matrix to the initial target exposure energy matrix to generate a registration compensation matrix. The core control unit reads the circuit design file to be processed, generates a binarized target image matrix, performs element-wise spatial Hadamard product operation on the binarized target image matrix and the registration compensation matrix, and outputs a composite grayscale matrix. The core control unit sends the composite grayscale matrix to the DMD digital micromirror exposure engine. The DMD digital micromirror exposure engine includes a micromirror array, which includes micromirror units. The DMD digital micromirror exposure engine controls the flipping duty cycle of the micromirror units according to the composite grayscale matrix, and projects modulated exposure energy onto the copper foil surface coated with wet film. The solution in the developing tank dissolves and peels off the wet film in areas where no photocrosslinking reaction has occurred, forming the final conductive pattern mask.

Images