An assembly and adjustment method suitable for chromatographic volumetric additive manufacturing systems

By establishing a quantifiable alignment reference and hierarchical compensation mechanism in the VAM system, the problem of optical dose superposition distortion caused by mechanical assembly errors and dynamic jitter is solved, achieving high-precision 3D printing, reducing costs and simplifying the assembly process, and supporting the application of advanced materials and algorithms.

CN123104073APending Publication Date: 2026-11-17NORTHEAST NORMAL UNIVERSITY
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202610417195.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-04-01
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing volumetric additive manufacturing (VAM) systems, the problem of optical dose superposition distortion caused by mechanical assembly errors and dynamic jitter makes it difficult to achieve the requirements of geometric fidelity and dimensional accuracy, especially in high-precision application scenarios.

Method used

By establishing a quantifiable alignment reference and a graded compensation mechanism, including the establishment of a system coordinate system, projection lens pitch correction, resin container attitude pre-adjustment, alignment plane establishment, ideal rotation axis orientation calibration, and dynamic error compensation, precise alignment and compensation between the projection line and the rotation axis can be achieved.

Benefits of technology

It significantly improves printing accuracy, reduces hardware costs and assembly difficulty, provides a standardized assembly process, ensures high-fidelity, isotropic volumetric printing, and supports the application of advanced materials and algorithms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure FT_1
    Figure FT_1
  • Figure FT_2
    Figure FT_2
  • Figure FT_3
    Figure FT_3
Patent Text Reader

Abstract

This invention discloses an assembly and adjustment method suitable for volumetric additive manufacturing (VAM) systems. It establishes an alignment plane where the image plane of the projection module and the object plane of the camera coincide as a unified detection benchmark. Within this plane, it achieves precise calibration and hierarchical compensation of the spatial relationship between the projection lines and the rotation axis to solve the problem of spatial superposition distortion of light dose caused by neglecting mechanical alignment in existing VAM technologies. Specifically, it includes: providing a geometric benchmark through leveling and establishing a coordinate system using a turntable (4); correcting the pitch of the projection module (6) using a near aperture (7a) and a far aperture (7b) and a laser (8); adjusting the posture of the resin container (15) using a five-degree-of-freedom adjuster (9) to ensure it is located in the effective printing area; obtaining the ideal rotation axis orientation within the alignment plane (18) through multi-cycle contour fitting; projecting the center column stripes to obtain the projection lines (17), achieving static alignment through coarse adjustment in the Y-axis direction and software geometric transformation based on fixed residual deviations; and further implementing angle-related dynamic compensation by analyzing the periodic residuals during rotation. This invention shifts the high-precision alignment requirement from expensive mechanical manufacturing to repeatable digital compensation, significantly reducing system hardware costs and assembly complexity. This method provides a stable and reliable physical alignment foundation for VAM systems and can be used in conjunction with techniques such as material refractive index matching and digital pre-compensation for optical distortion. It is particularly suitable for volumetric additive manufacturing of precision parts with extremely high geometric fidelity requirements.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of 3D printing technology, and more particularly to an assembly and adjustment method suitable for volumetric additive manufacturing (VAM) systems. This method solves the problem of optical dose superposition distortion caused by mechanical assembly errors and dynamic jitter by establishing a quantifiable alignment reference and a hierarchical compensation mechanism. It is particularly suitable for applications requiring extremely high geometric fidelity and dimensional accuracy, such as the printing of optical components, ceramic structures, or precision glass devices (e.g., fused silica hemispherical resonators). Background Technology

[0002] Volumetric additive manufacturing (VAM) achieves the integral molding of three-dimensional structures by projecting calculated two-dimensional images from multiple angles onto a rotating photosensitive resin container (15). Its molding accuracy fundamentally depends on the geometric alignment accuracy between the projection module (6) and the rotation axis, as well as the accurate transmission of the projected light in the complex optical path.

[0003] Currently, technological improvements to enhance VAM accuracy mainly fall into three categories, but all of them suffer from structural defects:

[0004] Digital pre-compensation methods for projected patterns (such as CN114302802B): By simulating the propagation path of light in the container (15) and resin, digital pre-compensation is performed on the projected pattern to counteract optical distortion caused by the shape or material refraction of the container (15). However, this method completely ignores systematic geometric alignment errors caused by mechanical assembly deviations (such as installation errors of the projection module (6), eccentricity and wobbling of the turntable (4)). Its assumption is that the projection module (6) and the rotation axis are perfectly aligned, which is seriously inconsistent with engineering reality.

[0005] Material optimization methods for scattering media (such as CN114269544A) reduce light scattering by adjusting the refractive index of the resin matrix to match suspended particles (such as cells or ceramic powders). However, this method only addresses the optical homogeneity of the material itself and cannot resolve spatial dose distribution errors caused by projection path distortion and mechanical misalignment.

[0006] Iterative optimization methods based on reconstruction quality assessment (such as CN116681841A) optimize the input image by evaluating the edge error of the tomographic reconstruction model. This method focuses on backend data processing and algorithm iteration, and fails to provide a quantifiable benchmark method and compensation mechanism that can achieve high-precision alignment in the physical assembly stage. Moreover, the optimization process is computationally intensive and inefficient.

[0007] The closest and most dangerous existing technology is the first type of method mentioned above (CN114302802B). Its structural defect is that it is only a "software correction" method, and its effectiveness depends heavily on an unrealistic "ideal hardware platform" assumption. In actual VAM systems, the mechanical errors introduced by the rotation accuracy of the turntable (4), the installation tolerance of the projection module (6), and the concentricity of the container (15) are often on the order of millimeters or even hundreds of micrometers, far exceeding the optical aberrations it attempts to correct. The direct consequence of ignoring these mechanical errors is that even if perfect ray tracing compensation is performed in the software, the actual projected light pattern will still experience "misaligned exposure" due to the non-coincidence of the projection line (17) with the rotation axis, resulting in dimensional deviations, shape distortions, and severe anisotropy in the final formed structure.

[0008] Therefore, the core technical problem to be solved by this invention is to provide a quantifiable, repeatable alignment and adjustment method that can be implemented in the physical assembly stage of the VAM system and does not rely on extremely high-cost mechanical precision. This method needs to be able to directly detect and compensate for the geometric deviation between the projection line (17) and the rotation axis, and further eliminate the dynamic error during the rotation process, thereby ensuring the accurate superposition of light dose in three-dimensional space at the source, providing a reliable and aligned physical basis for any subsequent advanced image algorithm compensation (such as optical distortion compensation). Summary of the Invention

[0009] Purpose of the Invention: The purpose of this invention is to overcome the systematic precision defects in existing VAM technology caused by neglecting mechanical alignment, and to provide a complete assembly and adjustment method. This method aims to:

[0010] Establish a quantifiable alignment plane (18) and a detection reference to realize the direct measurement of the spatial positional relationship between the projection line (17) and the rotation axis.

[0011] By employing a tiered strategy of "static hardware coarse adjustment + static software geometric compensation + dynamic angle-related correction," the high-precision alignment requirement is shifted from expensive mechanical manufacturing to repeatable digital compensation, significantly reducing system hardware costs and assembly difficulty.

[0012] A standardized and engineered assembly and commissioning process has been developed, including clear criteria, operating procedures, and exception handling, to ensure the high repeatability and stability of the method.

[0013] By combining this method with techniques such as material optical optimization and digital pre-compensation of projected images, we verify its effectiveness and necessity in ultra-high precision manufacturing tasks (such as printing fused silica glass devices), and ultimately achieve high-fidelity, isotropic volumetric printing of complex structures.

[0014] Technical Solution: To achieve the above objectives, this invention proposes an assembly and adjustment method suitable for tomographic volumetric additive manufacturing systems. The method uses an alignment plane (18) where the image plane of the projection module (6) and the object plane of the camera (12) coincide as a unified spatial reference, and performs detection and compensation of all geometric relationships within this plane. The overall process of the method is as follows: Figure 9 As shown, it includes the following core steps:

[0015] S1: Establishing the system coordinate system and leveling the turntable

[0016] The system turntable (4) is leveled using a high-precision level (3), and the rotation axis of the turntable (4) is defined as the Z-axis. The normal direction of the ideal image plane of the projection module (6) is defined as the X-axis, and the Y-axis is determined according to the right-hand rule, thus establishing the system's rectangular coordinate system. This step provides a unified and accurate geometric reference framework for the description and calculation of all subsequent spatial relationships. The operation of this step is illustrated as follows: Figure 1 As shown.

[0017] (Solving sub-problems: Establishing an absolute benchmark for spatial measurements and eliminating errors introduced by the tilt of the installation foundation.)

[0018] S2: Projector lens tilt correction

[0019] In front of the lens of the projection module (6), a near aperture (7a) and a far aperture (7b) are set sequentially along the X-axis. A laser (8) with a precision exit aperture is set on the opposite side of the projection module (6). The attitude of the laser (8) is adjusted so that the laser beam passes through the near aperture (7a) and the far aperture (7b) sequentially and enters the projection lens, achieving preliminary collinearity of the optical path. Subsequently, a collimating reference light is emitted from the projection lens, and the pitch angle of the projection module (6) is finely adjusted based on whether the reference light can be accurately reflected back into the exit aperture of the laser (8) as the final criterion until the reflection condition is met. This step ensures that the projected image plane is strictly parallel to the YZ plane of the system, eliminating the trapezoidal distortion of the image caused by the tilt of the projection plane. The principle of adjusting the optical path is as follows: Figure 2 As shown.

[0020] (Solve the sub-problem: ensure that the projected image does not undergo inherent geometric deformation during transmission.)

[0021] S3: Resin container (bottle) posture pre-adjustment and tolerance control

[0022] A cylindrical container (vial) (15) containing photosensitive resin is mounted on the center of a turntable (4) using a fine-tuning fixture (9) with five degrees of freedom (three translations and two rotations). The knob of the fixture (9) is adjusted so that the lateral offset (eccentricity) and the included angle (tilt) of the geometric axis of the vial (15) relative to the rotation axis (Z-axis) of the turntable (4) are controlled within a preset tolerance range (e.g., eccentricity < 0.4 mm, tilt < 0.5°).

[0023] This operation serves two purposes: firstly, to ensure that the target printing volume inside the vial (15) remains within the "effective printing area" that can be illuminated from all projection angles when the vial (15) rotates around the Z-axis; secondly, it should be understood that although the position of the vial (15) does not need to be perfectly concentric, the eccentricity should not be too large. The reason is that during VAM printing, the projection beam usually needs to pass through the refractive index matching liquid and the vial (15) before entering the resin medium. When the axis of the vial (15) is eccentric or tilted, the actual propagation path of the beam inside the refractive index matching liquid and the vial (15) will change with the rotation angle, thereby introducing angle-related refractive offset and equivalent projection position drift, causing a systematic deviation between the actual projection pattern and the ideal projection pattern, which in turn affects the accuracy of the three-dimensional light dose superposition and the geometric fidelity of the forming. Therefore, this step adjusts and locks the axis of the vial (15) by setting and using the five-degree-of-freedom fine-tuning fixture (9) to constrain the eccentricity / tilt within the tolerance range, stabilize the optical path conditions during the printing process, and provide a reliable premise for subsequent alignment plane calibration and static / dynamic compensation. Adjustment diagram as shown Figure 3 As shown.

[0024] (Solve the sub-problem: avoid printing defects caused by some structures "moving out" of the light field during rotation due to improper installation of the container (15); at the same time, suppress the angle-related light path changes caused by the eccentricity of the small bottle axis, and reduce the deviation of the actual projected graphic from the ideal graphic.)

[0025] S4: Establishing the alignment plane (X-axis alignment)

[0026] A camera (12) is arranged on the opposite side of the projection module (6), and a uniformly emitting LED light-emitting panel (14) is set behind the small bottle (15) in the optical axis direction of the camera (12).

[0027] a) Camera position locking: Turn off the projection module (6), turn on the LED light panel (14), and capture the backlit outline image of the bottle (15) while the turntable (4) is stationary. Adjust the position of the camera (12) along the X-axis, using the narrowest line width and the sharpest outline of the left and right edges of the bottle (15) in the image captured by the camera (12) as the criterion. At this time, the object plane of the camera (12) coincides with the center plane of the container (15), and the position of the camera (12) is locked. This step, by moving the camera (12) along the X-axis, makes its object plane precisely coincide with the center plane of the container—establishing a clear imaging reference for subsequent image acquisition.

[0028] b) Projection Module Position Locking: Keep the camera (12) in a fixed position, turn off the LED light panel (14), turn on the projection module (6) and project a preset thin line test pattern. Adjust the position of the projection module (6) along the X-axis, and lock the position of the projection module (6) based on the criterion that the line width of the test pattern in the image of the camera (12) is the narrowest and the outline is the clearest. This step moves the projection module (6) along the X-axis so that its image plane coincides with the object plane of the camera (12) conjugately—unifying the projection imaging plane with the visual inspection plane, establishing a quantifiable alignment plane (18), and providing a common reference benchmark for all subsequent geometric measurements.

[0029] Through this step, the image plane of the projection module (6) and the object plane of the camera (12) achieve conjugate coincidence, and this coincident plane is defined as the "alignment plane (18)". All subsequent two-dimensional geometric measurements and compensations regarding the "projection line (17)" and the "rotation axis" are performed within this plane. The principle of this X-axis direction adjustment is as follows: Figure 4 As shown.

[0030] (Solving the subproblem: Create a physically detectable two-dimensional reference plane that is directly related to projection imaging and visual inspection, simplifying the three-dimensional alignment problem into a geometric problem within the plane.)

[0031] S5: Calibration of the ideal rotation axis orientation (multi-period fitting)

[0032] Under LED illumination (14), the vial (15) is driven to rotate continuously at a constant angular velocity (e.g., 24° / s). The camera (12) acquires a sequence of contour images for multiple complete rotation cycles (e.g., lasting 3 minutes). Each frame is processed using an image processing program as follows:

[0033] The edge detection algorithm was used to extract the point set of the left and right outer edges of the vial (15).

[0034] By fitting straight lines to the left and right edge point sets respectively, two edge lines are obtained.

[0035] Calculate the geometric midline of these two lines. This midline is the projection line of the geometric axis of the small bottle (15) onto the alignment plane (18) at the current rotation angle.

[0036] Statistical analysis was performed on the "projected straight line of the bottle (15) axis" for all cycles and angles. The least squares method or a sine model was used for fitting to obtain an optimal fitted straight line. This straight line represents the precise projection orientation of the turntable (4) rotation axis within the alignment plane (18) after eliminating the bottle (15)'s own installation errors and random jitter, and is defined as the "ideal rotation axis". The fitting process and data examples are as follows: Figure 7 As shown.

[0037] (Solve the sub-problem: accurately measure the true position and orientation of the rotation axis in the reference plane, as the ultimate target benchmark for all alignment.)

[0038] S6: Coarse adjustment of the Y-axis direction of the projection line and software compensation for fixed residual deviation

[0039] The projection module (6) projects a high-contrast single-column stripe pattern ("center column stripe") generated by two adjacent columns of center pixels onto the alignment plane (18). The camera (12) acquires the stripe image and fits the line connecting the brightness centroids of each row of stripes into a straight line, which is defined as the "projection line (17)".

[0040] Y-axis coarse adjustment: Within the field of view of the camera (12) aligned with the plane (18), superimpose the "projection line (17)" and the "ideal rotation axis" obtained in step S5. Manually adjust the position of the projection module (6) along the Y-axis and observe in real time until the angle and lateral distance (Y-axis deviation) between the two lines in the plane are less than the preset coarse adjustment threshold (e.g., angle < 0.1°, deviation < 5 pixels). Lock the position of the projection module (6) along the Y-axis. This adjustment is illustrated as follows. Figure 7 As shown. This step, by adjusting the position of the projection module (6) along the Y-axis, achieves the initial alignment of the projection line (17) with the ideal rotation axis, providing a good initial state for subsequent fine compensation.

[0041] S7: Fixed Residual Deviation Measurement and Software Compensation: After coarse adjustment, under stable rotation of the system, multi-angle contour images are acquired again to accurately calculate the residual fixed deviation of the "ideal rotation axis" relative to the "projection line (17)", including a translation (ΔY) perpendicular to the axis and a small angle rotation (Δθ). Subsequently, when generating the final two-dimensional projection image set for printing, a uniform translation and rotation transformation with the same value and opposite to the above residual deviation is applied to the entire image set. This operation is completed in the digital domain, which is equivalent to using software fine-tuning to make the "projection line (17)" and the "ideal rotation axis" completely coincide in the plane based on hardware alignment.

[0042] (Solving sub-problems: 1. Achieve rapid preliminary alignment through hardware; 2. Compensate for micron-level residual errors that cannot be eliminated by hardware through software geometric transformation, achieving sub-pixel-level alignment accuracy.)

[0043] S8: Regularity Analysis of Dynamic Errors and Final Compensation Based on Angle Correlation

[0044] After completing the static compensation in step S6, the system may still have dynamic errors related to the rotation angle caused by periodic runout of the turntable (4) bearing, uneven transmission, etc.

[0045] Data acquisition: Continue to record the real-time deviation (displacement residual, angle residual) between the "ideal rotation axis" and the "equivalent projection line (17)" after static compensation within multiple rotation cycles. Some typical data are as follows: Figure 8 As shown.

[0046] Pattern analysis: By analyzing the curve of residual data changing with rotation angle, it is usually possible to extract sine or more complex periodic components that are synchronized with the rotation period.

[0047] Generate a compensation table: Convert the angle-residual relationship obtained from the analysis into an angle-compensation lookup table (LUT), such as... Figure 8 As shown in the table below.

[0048] Final compensation: Based on the static compensation, according to the current projection angle, the corresponding additional compensation amount (a small translation or rotation) is read from the LUT and applied to the projected image corresponding to that angle in real time. This step achieves the suppression of the system's periodic dynamic jitter.

[0049] (Solve the sub-problem: compensate for the periodic error introduced by the imperfect motion of the turntable (4), and further improve the instantaneous accuracy of the light dose superposition, which is especially important for structures with high aspect ratio or extremely high requirements for symmetry.)

[0050] Collaborative Application and Verification: To verify the effectiveness of the assembly and adjustment method described in this invention and its application value in high-end manufacturing, it can be combined with the following technologies for application to challenging printing tasks, such as the fabrication of fused silica hemispherical harmonic oscillators:

[0051] Synergistic with Material Optics Optimization: In the assembled and adjusted VAM system, a liquid glass material optimized for refractive index matching (such as a composite resin containing HEMA, TEGDA, PC, DEDB, and nano-SiO2) is used. The precise alignment of this invention ensures accurate positioning of the projected light rays, while the low-scattering material guarantees an accurate attenuation model of light intensity within the resin. The combination of these two aspects forms the basis for achieving high-fidelity printing of complex glass structures.

[0052] Collaboration with digital distortion pre-compensation: On systems with achieved geometric alignment, ray tracing algorithms such as those described in CN114302802B can be further applied to pre-compensate for optical distortion caused by the walls of the cylindrical container (15). This invention provides a reliable physical platform for such pure software compensation, avoiding the logical dilemma of "performing distortion compensation on a distorted hardware basis".

[0053] Synergistic with closed-loop optimization in the printing process: For resonator-like structures with thin walls, deep cavities, and high-precision rod systems, iterative optimization based on edge error assessment can be introduced during the printing process (similar to the idea in CN116681841A). The precise initial alignment of this invention greatly reduces the number and magnitude of iterations required for convergence, improving optimization efficiency. For example, when printing a hemispherical resonator, this method ensures that the projection axis coincides with the rotation axis, fundamentally reducing systematic errors such as uneven shell thickness and rod eccentricity. Subsequent adjustments only require fine-tuning of small-scale material curing threshold variations.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] To fundamentally improve printing accuracy: by directly aligning the projection line (17) with the physical rotation axis, the biggest systematic error source caused by mechanical misalignment is eliminated from the source, laying a physical foundation for high-fidelity printing.

[0056] Significantly reduce hardware costs and assembly difficulty: The innovative adoption of a graded compensation strategy of "hardware coarse adjustment + software fine adjustment" shifts the stringent precision requirements from difficult-to-achieve mechanical manufacturing tolerances to flexible and repeatable digital compensation, allowing the use of lower-cost turntables (4) and moving parts, while simplifying the assembly process.

[0057] Standardized and reusable engineering methods are formed: a complete assembly and commissioning process with clear criteria, operating procedures and exception handling logic is provided, which makes equipment assembly, container (15) replacement and periodic calibration fast and easy, greatly improving the maintainability, stability and repeatability of the VAM system among different operators.

[0058] Possessing strong technological synergy and application scalability: The precise and stable physical alignment platform provided by this invention is a necessary prerequisite and enabling technology for successfully applying advanced materials (such as low-scattering composite resins) and advanced algorithms (such as optical distortion pre-compensation and artificial intelligence iterative optimization) to further improve VAM performance. Its successful application in the fabrication of ultra-high precision parts such as fused silica harmonic oscillators strongly demonstrates the technological advancement and broad application prospects of this method. Attached Figure Description

[0059] Figure 1 A schematic diagram for leveling the turntable (4) and establishing the coordinate system;

[0060] Figure 2 A schematic diagram for adjusting the pitch of the projection module (6) lens, with the near aperture (7a) and the far aperture (7b) marked in the diagram.

[0061] Figure 3 A schematic diagram is created to adjust the position of the projection module (6) and the alignment plane (18) in the X-axis direction;

[0062] Figure 4 A schematic diagram illustrating edge detection and fitting of a vial, with the rotation axis position calibrated on the camera screen.

[0063] Figure 5 A schematic diagram showing how the position of the projection module is adjusted along the y-axis to make the projection line coincide with the ideal rotation axis.

[0064] Figure 6 A schematic diagram showing how adjusting the five-degree-of-freedom regulator can achieve lateral offset and included angle of the vial shaft relative to the rotation axis within the tolerance range;

[0065] Figure 7 This is a schematic representation of dynamic error acquisition, analysis, and angle-related compensation.

[0066] Figure 8 A schematic diagram and data fitting curve for obtaining the ideal rotation axis through multi-period fitting;

[0067] Figure 9 This is an overall flowchart of the assembly and adjustment method described in this invention;

[0068] In the diagram: 1. Front pivot 2. Rear pivot 3. Level 4. Turntable 5. Turntable support 6. Projection module 7a. Near aperture 7b. Far aperture 8. Laser 9. Five-degree-of-freedom adjuster 10. Projector image plane or camera object plane (not aligned) 11. Test tube clamp 12. Camera 13. Projector image plane or camera object plane (aligned) 14. LED light panel 15. Resin bottle 16. Square glass box 17. Projection lines 18. Camera screen (aligned to the plane) Detailed Implementation

[0069] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings and examples. These examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0070] Example 1: Baseline Assembly and Alignment Process

[0071] This embodiment demonstrates the entire process of initial assembly and alignment of a typical VAM system using the method of this invention. The overall steps of the method are as follows: Figure 9 As shown.

[0072] Hardware deployment: Install the turntable (4), projection module (6) (DLP projector), camera (12), laser (8), near aperture (7a) and far aperture (7b), LED light panel (14) and five-degree-of-freedom container fixture (9) according to the system layout.

[0073] Perform S1: Use an electronic level (3) (accuracy 0.01°) to adjust the turntable support (5) so that the plane of the turntable (4) is horizontal, with an error ε ≤ 0.02°. Establish a coordinate system with the rotation axis as the Z-axis and the projection direction as the X-axis. This operation is illustrated as follows. Figure 1 As shown.

[0074] Execute S2: such as Figure 2 As shown, a near aperture (7a) and a far aperture (7b) with a spacing of approximately 50cm are installed (aperture diameter 1mm). Adjust the laser (8) so that the beam passes through the center of the near aperture (7a) and the far aperture (7b) and enters the center of the lens of the projection module (6). Switch to the light output mode of the projection module (6) and fine-tune the tilt knob of the projection module (6) until the collimated light spot emitted can be reflected back into the output aperture (aperture diameter 2mm) of the laser (8). Lock the projection module (6).

[0075] Execute S3: such as Figure 3 As shown, a cylindrical glass bottle (15) (20 mm in diameter) containing a refractive index matching solution is placed into a five-degree-of-freedom fixture (9). In a static state, the outline of the bottle is observed through a camera (12). The XY translation and tilt knobs around X and Y of the fixture (9) are adjusted to make the outline of the bottle symmetrically centered, with an eccentricity δ_bottle ≤ 0.5 mm and a tilt α_bottle ≤ 0.5°.

[0076] Execute S4: such as Figure 4 As shown, turn on the LED light panel (14) and the camera (12) captures the static outline of the bottle. Move the camera (12) along the guide rail and observe the clarity of the left and right edges of the bottle in the image. When the edge line width is the narrowest (about 2-3 pixels) and the grayscale gradient is the steepest, lock the camera (12). This step, by moving the camera (12) along the X-axis, makes its object plane precisely coincide with the center plane of the container—establishing a clear imaging reference for subsequent image acquisition. Turn off the LED light panel (14) and the projection module (6) projects a set of vertical line patterns. Move the projection module (6) and lock its position based on the narrowest line width of the pattern. This step, by moving the projection module (6) along the X-axis, makes its image plane conjugate to the object plane of the camera (12)—unifying the projected imaging plane with the visual detection plane and establishing a quantifiable alignment plane (18). At this time, the image plane of the projection module (6) is conjugate with the object plane of the camera (12), and the alignment plane (18) is established.

[0077] Execute S5: such as Figure 5As shown, the LED light panel (14) is turned on, and the turntable (4) is driven at ω = 24° / s. The camera (12) continuously acquires video for 180 seconds (approximately 3 cycles) at 30fps. The program automatically processes each frame: Canny edge detection within the ROI, fitting the left and right edge lines, and calculating the midline. All midline data are archived by angle. Figure 7 As shown, a sine fit is performed on the position y(θ) of the centerline in the Y-axis direction: y(θ) = A sin(θ + φ) + y0, resulting in the fitted curve. The centerline y = y0 of this curve represents the horizontal position of the ideal rotation axis in the alignment plane (18), and its direction is determined by the fitting parameters.

[0078] Perform S6: Project the center column stripes (2 pixels wide), and the camera (12) acquires and fits the projection lines (17). For example... Figure 6 As shown in the left part, the projection line (17) (green) and the ideal rotation axis (red dashed line) are superimposed on the software interface. An angle and offset between the two are observed. Loosen the Y-axis locking mechanism of the projection module (6), manually adjust its position, and observe the software interface until the two lines are basically aligned (angle < 0.05°, Y-axis deviation < 3 pixels), then relock. This step achieves the initial alignment of the projection line (17) with the ideal rotation axis by adjusting the position of the projection module (6) along the Y-axis. This completes the coarse adjustment.

[0079] Execute S7: Subsequently, the system acquires 100 more contour images while in rotation, and calculates the average residual deviation: ΔY = -0.8 pixels, Δθ = +0.008°. In subsequent printing, all projected images are uniformly transformed by translation (+0.8 pixels) and rotation (-0.008°) after loading before being projected.

[0080] Execute S8: Keep the system running and continue collecting residual data (δ_i, α_i) for 500 consecutive angles. Analysis reveals a clear regularity in the residuals with a period of 360°. Averaging the residuals within each angle interval (e.g., every 5° interval) generates an angle-compensation scale (LUT) containing 72 items, with a similar data format. Figure 8 As shown. This LUT is integrated into the projection control software. In subsequent printing, when the projection angle is θ, the software not only applies the fixed compensation of S6, but also queries and applies additional dynamic compensation amounts (Δδ_θ, Δα_θ) from the LUT.

[0081] Example 2: Application Verification in the Fabrication of Fused Quartz Hemispherical Harmonic Oscillators

[0082] This embodiment demonstrates the use of the VAM system aligned with Embodiment 1 for printing high-precision fused silica hemispherical harmonic oscillators and showcases its synergy with other technologies.

[0083] Material Preparation: A liquid glass material with a matching refractive index was prepared. Components: HEMA (28.1 wt%), TEGDA (3.7 wt%), PC (8.9 wt%), DEDB (5.6 wt%), nano-SiO2 (Ox50, 53.7 wt%), photoinitiator 819 (0.014 wt%), co-initiator EDAB (0.1 wt%), and photoinhibitor TEMPO (0.0014 wt%). Measurements showed that the refractive index of the organic phase, n_org ≈ 1.458, matched well with SiO2 (n ≈ 1.46), and the material exhibited a transmittance >85% at 405 nm wavelength (optical path 10 mm).

[0084] Digital pre-compensation: Based on the precisely aligned physical system of Example 1, the light propagation matrix of the cylindrical glass bottle (15) (wall thickness 1.5 mm, refractive index 1.47) and the aforementioned liquid resin (refractive index ≈ 1.46) was measured. Using this matrix, pre-distortion calculations were performed on the original Radon transform projection image of the target hemispherical harmonic oscillator (rod diameter 1.6 mm, shell thickness 0.5 mm) to generate a compensated image set.

[0085] Printing and initial observation: The prepared material was injected into the bottle (15) and printing was performed. The projected light intensity was 0.8 mW / cm², and the rotation speed was 24° / min. By monitoring the CCD, it could be observed that the resonator structure (rod and shell) began to appear at multiple angles almost simultaneously, and the curing of the shell started from multiple points rather than from top to bottom, indicating that the light dose distribution was uniform and the alignment was good.

[0086] Closed-loop optimization (optional): A microscopic CT scan was performed on the initially printed gel-state resonator to reconstruct the 3D model. The edge region at the rod-shell connection was extracted, and its edge error (EdgeE) was calculated (method referred to CN116681841A). Slight over-curing was found at the rod root (EdgeE = 8%). Based on this, the grayscale of the rod region at the corresponding angle in the projected image set was fine-tuned (reduced by approximately 5%). After reprinting, EdgeE decreased to below 3%.

[0087] Post-processing and characterization: After cleaning and drying according to standard procedures, the printed parts underwent heat treatment (degreasing at 600℃ and sintering at 1180℃). The final fused silica hemispherical harmonic oscillator was transparent and crack-free. Three-dimensional morphology measurements showed: rod diameter deviation < ±25μm, shell thickness non-uniformity < 8%, and shell roundness error < 20 μm. Its accuracy is significantly better than the control group that was directly printed and processed without using the alignment method of this invention.

[0088] Exception handling and boundary conditions:

[0089] If the goodness of fit R² of S5 is < 0.95, it indicates "poor rotational stability or loose installation of container (15). It is recommended to check the turntable (4) drive, the clamp (9) locking and the concentricity of container (15), and re-execute S3-S5.

[0090] If the S7 dynamic residual amplitude is too large (e.g., > 10 pixels): the message "There is a serious periodic fault in the mechanical transmission" will be displayed. It is recommended to perform mechanical maintenance instead of relying solely on software compensation.

[0091] After changing to a container of different diameter / material (15): S4 (to confirm the alignment plane (18)) and S5 (to update the ideal rotation axis) need to be re-executed, but the fixed compensation parameters of S6 and the dynamic compensation table of S7 can usually be used or only need to be finely adjusted, greatly shortening the changeover time.

[0092] 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 or procedural transformations made based on the inventive concept of the present invention and the contents of the specification and drawings of the present invention, or direct / indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. An assembly and adjustment method suitable for chromatographic volumetric additive manufacturing systems, characterized in that, Includes the following steps: S1. Establishment of system coordinate system and leveling of turntable: Level the system turntable (4) using a level (3) to establish a system rectangular coordinate system with the rotation axis as the Z-axis and the normal of the ideal image plane of the projection module (6) as the X-axis; S2. Projection lens pitch correction: By adjusting the collinearity of the laser (8) with the near aperture (7a) and the far aperture (7b), the pitch of the projection module (6) is finely adjusted so that the projection image plane is parallel to the YZ plane; S3. Pre-adjustment of resin container posture: Install the resin container (15) through a five-degree-of-freedom adjuster (9) to control the lateral offset and included angle of its bottle axis relative to the rotation axis within the preset tolerance range; S4. Alignment plane establishment: A camera (12) and an LED light-emitting plate (14) are arranged on the opposite side of the projection module (6). The image plane of the projection module (6) and the object plane of the camera (12) are conjugate and coincident by adjusting the X-axis direction. This conjugate plane is defined as the alignment plane (18). S5. Ideal rotation axis calibration: drive the container (15) to rotate and acquire multi-cycle backlight contour images. Obtain the projection line of the container axis in the alignment plane (18) at each angle through edge detection and fitting, and obtain the ideal rotation axis that represents the best orientation of the rotation axis by fitting the multi-cycle data. S6. Projection line alignment and static compensation: Project the center column stripes onto the alignment plane (18) and fit the projection line (17). Adjust the projection module (6) along the Y-axis direction so that the projection line (17) initially coincides with the ideal rotation axis. S7. Measure the fixed residual bias and apply reverse translation and rotation transformations to all projected images for software compensation; S8. Dynamic Error Compensation: Based on static compensation, multi-cycle dynamic residual data is collected, the regularity of its variation with rotation angle is analyzed, an angle-compensation lookup table is generated, and final dynamic compensation is applied to the projected image at a specific angle.

2. The method according to claim 1, characterized in that, In step S2, the apertures of the near aperture (7a) and the far aperture (7b) are both small aperture structures. The distance between the far aperture (7b) and the lens of the projection module (6) is much greater than the distance between the near aperture (7a) and the lens of the projection module (6) to amplify the pitch deviation of the projection module (6). The adjustment optical path is shown in Figure 2.

3. The method according to claim 1, characterized in that, In step S3, the resin container (15) is installed on the turntable (4) by a vial shaft adjustment device. The vial shaft adjustment device is a five-degree-of-freedom adjuster (9) with two-way translation adjustment and three-way tilt adjustment. By adjusting the adjuster (9), the eccentricity δ_vial and tilt angle α_vial of the geometric axis of the resin container (15) relative to the rotation axis are limited within a preset tolerance range, so as to reduce the dose deviation caused by the change of the propagation path of the projected light in the refractive index matching solution and the resin container (15) with the rotation angle due to the eccentricity of the vial shaft.

4. The method according to claim 1, characterized in that, The acquisition of the ideal rotation axis in step S5 includes: performing edge detection on each frame image to obtain the left and right edge lines of the container, calculating the geometric midline of the two lines as the container axis projection at the current angle; performing sine model or least squares fitting on the container axis projection data at all angles within multiple complete rotation cycles to obtain the best-fit line, and an example of its data processing is shown in Figure 6.

5. The method according to claim 1, characterized in that, The fixed residual deviation in step S7 includes a translation ΔY perpendicular to the ideal rotation axis and a small-angle rotation Δθ about the normal of the alignment plane (18); the software compensation is to apply a composite geometric transformation of translation and rotation of size (-ΔY, -Δθ) to all projected images used for printing.

6. The method according to claim 1, characterized in that, The method for generating the angle-compensation lookup table in step S8 is as follows: record the displacement residual δi and angle residual α_i corresponding to each sampling angle θi within N complete rotation cycles; calculate the statistical average or fitted value of the residual sequence corresponding to each θi to form a mapping relationship LUT: θi → (Δδi, Δα_i), and its data representation is shown in Figure 8.

7. The method according to claim 1, characterized in that, The method further includes a step of synergistic application with materials optical optimization technology and digital distortion pre-compensation technology: a) Printing is performed using a photosensitive material optimized for refractive index matching in the VAM system after the assembly and adjustment are completed; b) Based on the alignment, a digital pre-compensation algorithm based on ray tracing or propagation matrix is ​​applied to correct the optical distortion caused by the container wall and resin; Among them, the precise mechanical alignment provided by steps S1-S8 provides a reliable physical basis for the digital pre-compensation algorithm in step b).

8. The method according to claim 6, characterized in that, The photosensitive material optimized for refractive index matching includes a liquid glass material for manufacturing fused silica glass devices, wherein the liquid glass material comprises an organic monomer, a pore-forming agent, and nano-silica particles, and the difference between the refractive index of the organic phase and the refractive index of the silica is less than 0.

02.

9. The method according to claim 1, characterized in that, When applied to the volumetric additive manufacturing of fused silica hemispherical harmonic oscillators, the method also includes a closed-loop optimization step in the printing process: based on the edge region error assessment results of the printed structure, the grayscale or geometric distribution of the corresponding region in the projected image set is iteratively adjusted until the edge error is lower than a set threshold.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program performs the image acquisition, data processing, deviation calculation, and compensation parameter generation functions involved in steps S4 to S8 of the method according to any one of claims 1 to 9.

11. A chromatographic volumetric additive manufacturing system, characterized in that, It includes a turntable (4), a projection module (6), a camera (12), a processing unit, and a memory, wherein the memory stores a computer program, and the processing unit executes the program to control the system to implement the method of any one of claims 1 to 9.

Citation Information

Patent Citations

  • Method and apparatus for volumetric additive manufacturing of cell-loading resin

    CN114269544A

  • Volume Additive Manufacturing Method and Equipment with Digital Distortion Compensation

    CN114302802B

  • Fault reconstruction quality evaluation method and storage medium

    CN116681841A