A reverse compensation method for a cavity of an optical-grade injection mold

Abstract

This invention discloses a method for reverse compensation of the cavity in an injection mold for optical-grade parts, belonging to the field of precision injection molding and mold design technology. The method involves acquiring a standard surface model of the target optical-grade injection molded part and discretizing it to establish a standard point cloud with node numbers. Combining molding simulation, the method obtains the deformation vector of each node and calculates the specific volume change of each node from the reference state to the target state based on the material PVT model, obtaining the equivalent amplification factor of the node. Using the deformation vector as the compensation direction, the method calculates the compensation amount according to the equivalent amplification factor, generating a non-uniform compensation point cloud. Further, the compensation point cloud is subjected to surface partitioning, surface fitting, and solid reconstruction to obtain a solid model of the compensation cavity. This invention can achieve non-uniform reverse compensation of the cavity in an injection mold for optical-grade parts, improving the surface accuracy of the part, reducing the number of trial moldings and repairs, and lowering R&D costs.

Description

Technical Field

[0001] This invention relates to the field of precision injection molding and mold design technology, specifically to a method for non-uniform reverse compensation of mold cavity to improve the surface accuracy of optical-grade injection molded parts. Background Technology

[0002] Injection molding is one of the most widely used processing techniques in the production of thermoplastic polymer products, offering advantages such as high molding efficiency, adaptability to complex structures, and ease of large-scale manufacturing. With the increasing demand for optical-grade injection molded parts such as automotive headlight lenses and imaging lenses, optical injection molding is gradually developing towards higher precision surface shapes and greater consistency. However, optical-grade injection molded parts are extremely sensitive to surface shape errors in the optical functional aperture areas. Non-uniform shrinkage and residual stress release generated by the material under the temperature-pressure coupling path during molding can cause surface shape errors, significantly affecting the optical performance of the product. In practical engineering, the non-uniform surface shape error of optical-grade injection molded parts is mainly formed by the combined effects of multiple factors, including the material's PVT characteristics, mold heat transfer and cooling structure, gate and pressure transmission, and process parameter settings.

[0003] To address these issues, the industry has developed various cavity compensation techniques. One type is experience-based uniform compensation, such as the single amplification factor used in traditional mold design or overall amplification based on material shrinkage rate. However, this method ignores the non-uniform shrinkage caused by historical differences in temperature and pressure in different areas of the part, resulting in limited compensation accuracy. Another type is simulation-based compensation methods, which obtain product shrinkage deformation through high-precision simulation and then perform non-uniform compensation on the mold cavity, significantly improving surface accuracy. However, this type of method usually relies on the shrinkage vector of the contour reference point and calculates the mold contour through geometric fitting, so its compensation basis is still the displacement field itself. In addition, some methods attempt to bridge the differences between simulation and actual measurement by introducing correction coefficients. Chinese patent CN114506103B proposes a method for inverting the design of the cavity of a small-module plastic gear injection mold. This method compares simulated shrinkage rates with experimental shrinkage rates to obtain correction factors for inverting the mold cavity design. Chinese patent CN114654623B proposes a method for correcting the deformation of injection molds. This method establishes a correction coefficient K between the analyzed deformation and the measured value through an initial mold trial, which is used to predict the compensation effect of the secondary deformation model. While these methods improve the accuracy of compensation to some extent, their correction process relies on feedback from physical mold trials, and the correction coefficient is not directly related to the physical properties of the material (PVT characteristics). This results in a long iteration cycle and high cost.

[0004] Therefore, there is still a lack of a compensation method in the existing technology that starts directly from the physical essence of the material shrinkage characteristics, rather than simply relying on reverse displacement or empirical correction, that can solve the common technical problems of unclear physical basis and difficulty in characterizing non-uniform shrinkage in existing compensation methods, and can accurately characterize and compensate for local non-uniform shrinkage differences in the simulation stage. Summary of the Invention

[0005] This invention proposes a reverse compensation method for the cavity of an injection mold for optical-grade parts. By introducing a PVT characteristic model of the material itself, the method uses the change in specific volume of the node at the end of the holding pressure and at room temperature as the physical scale basis for calculating the equivalent amplification factor, rather than simply reverse displacement or relying on empirical correction. This fundamentally solves the common technical problems of unclear physical basis and difficulty in characterizing non-uniform shrinkage in traditional compensation methods. It achieves true node-level non-uniform reverse compensation with physical connotation, improving the surface accuracy of optical-grade injection molded parts while reducing the cost of trial molding and mold repair.

[0006] The technical approach to achieve the above method is as follows: Standard surface point cloud coordinates, simulation-output node deformation vectors, and node temperature and pressure time-series data are associated with unified node serial numbers; the specific volume of the reference time and target state is obtained using the material PVT model, and the node volume ratio is calculated to obtain the equivalent amplification factor, thereby generating a scaled-down node compensation vector; the compensation vector is applied to the standard point cloud to generate a non-uniform compensation point cloud, and surface fitting and solidification are completed under closed boundary constraints and continuity constraints to output a compensation cavity solid model.

[0007] The method includes the following steps: Step S10: Standard surface shape definition and standard point cloud discretization modeling. Obtain the standard model of the optical-grade injection molded part and determine the optical functional aperture region; discretize the standard surface shape model to generate a simulation mesh, and output standard point cloud data X with unique node numbers and three-dimensional coordinate information. std，i (x,y,z); Extract feature surfaces and surface boundary information from the standard model; Using the standard point cloud 3D coordinates as a reference, associate the standard point cloud data node number with surface information and feature boundaries.

[0008] Step S20: Molding Simulation and Process Data Acquisition. Set molding process parameters, including but not limited to mold temperature, plasticizing temperature, injection speed, holding pressure, holding time, and cooling time; perform molding simulation under these parameters, and output node deformation vector data u consistent with the node number. i =(dx i ,dy i ,dz i ).

[0009] Step S30: Calculation of non-uniform compensation based on material PVT. The reference time is determined as the end of pressure holding or the termination of pressure transmission, and the target state is determined as room temperature and atmospheric pressure. At the reference time, the temperature and pressure of each node are extracted, and the specific volume v of each node at the reference time and the target state are obtained by combining the material PVT model. EOP,i v room,i Calculate the node volume ratio λ; calculate the equivalent magnification factor S of each node based on the node volume ratio. eff The node deformation vector is used as the compensation direction and the compensation amount is scaled according to the equivalent amplification factor to construct a node-scaled compensation vector.

[0010] Step S40: Non-uniform compensation point cloud generation and surface reconstruction preparation. Using the node index as a reference, the scaled compensation vector is applied to the standard point cloud to generate non-uniform compensation point cloud data X. cav，i The non-uniform compensation point cloud is used to characterize the geometric shape of the compensation cavity; the non-uniform compensation point cloud is divided into surface regions based on the node number, and the closed boundary point set of each surface region is extracted for subsequent surface fitting, trimming and splicing constraints.

[0011] Step S50: Constrained surface fitting and cavity solid generation. Under the constraint of closed feature boundary, the point set of each surface region is fitted to generate surface patches. Positional continuity and tangential continuity constraints are applied at the splicing of adjacent surface patches. The surface patches are then stitched and solidified to generate a compensated cavity solid model.

[0012] Furthermore, the material PVT model is obtained based on the isobaric heating path and is used to characterize the specific volume change relationship of the material under heating-pressurization conditions; the nodal volume ratio is determined by the specific volume ratio value between the reference time and the target state.

[0013] Furthermore, the non-uniform compensation point cloud in step S40 is obtained by applying a scaling compensation vector to each node of the standard point cloud based on the node number. The scaling compensation vector is the product of the equivalent magnification coefficient and the node deformation vector.

[0014] Preferably, the reference time is set to the end of the pressure holding period, at which point the gate is frozen and pressure transmission is cut off; the reference time can be determined by the gate flow rate-time curve, when the gate flow rate drops to a threshold of 1.0 × 10⁻⁶. - 3 When the pressure is below cc / s, pressure transmission is considered to have essentially ended, and this moment is determined as the reference moment.

[0015] Preferably, the standard point cloud data, node deformation vector data, and time-series data of node temperature and node pressure are all associated and read based on a unified node sequence number to realize the calculation of compensation amount and generation of compensation point cloud corresponding to each node.

[0016] Compared with the prior art, the present invention has the following beneficial effects: (1) The compensation amount has a physical scale basis. The specific volume of the reference time and the target state is obtained by using the material PVT model and the nodal volume ratio is calculated to obtain the equivalent amplification factor, thereby realizing the scaled compensation of the nodal deformation vector and being able to characterize the non-uniform shrinkage difference.

[0017] (2) Unified association of multi-source data and reproducibility at each node. Standard point cloud, node deformation vector and node temperature and pressure time series data are associated with a unified node number to realize the calculation of compensation amount and generation of compensation point cloud at each node.

[0018] (3) Improved surface accuracy and reduced cost. The method of the present invention can improve the surface accuracy of optical grade injection molded parts, reduce the number of trial molding and mold repair cycles, and has good engineering application value. Attached Figure Description

[0019] To more clearly illustrate the technical solution of the present invention, embodiments of the present invention will be further described in conjunction with the accompanying drawings, wherein: Figure 1 This is an overall flowchart of the method described in this invention; Figure 2 This is a schematic diagram of the three-dimensional structure of the optical lens used in the embodiment; Figure 3 This is a simulation model diagram of the optical lens used in the embodiment. Figure 4 For the example, the optical lens is simulated to show overall warping; Figure 5 Example: Flow rate-time curve of the simulated gate for the optical lens; Figure 6 This is an example of a solid model after non-uniform compensation; Figure 7 The following is a comparison diagram of the molding process of the three schemes in the embodiment. Detailed Implementation

[0020] To facilitate a further understanding of the technical solution of this invention, the invention will now be described in detail with reference to the accompanying drawings and embodiments. These embodiments are only used to illustrate the principles and implementation paths of this invention and are not intended to limit the scope of protection of this invention.

[0021] As shown in Figure 1, this invention proposes a reverse compensation method for the cavity of an injection mold for optical-grade parts based on material PVT.

[0022] Step S10: Standard surface definition and standard point cloud discretization modeling In this embodiment, an optical lens is used as the standard product. After measurement, its incident light surface curvature radius is 192.51 mm, the exit light surface curvature radius is 97.31 mm, and the diameter is 60 mm. The product model is shown in Figure 2.

[0023] In the standard model import phase, the lens solid model is meshed and used as the geometric basis for subsequent mesh discretization and physics calculations. This case study adopts a mesh control strategy tailored to optical accuracy requirements during the meshing phase. The mesh is locally refined in the effective optical aperture region, with a mesh size set to 100 μm. In non-critical areas, the mesh density is appropriately relaxed, with a mesh size set to 0.8 mm, to control the overall number of elements and shorten simulation time. The calculated mesh count on the product surface is 2,442,472.

[0024] The surface mesh was imported into the model flow software to generate a 3D solid mesh, with a total of 44,391,101 solid meshes. After the 3D solid mesh was generated using this method, a node database was created. Each node has a unique NodeID and its 3D coordinates are recorded. Node examples are shown in Table 1, with a total of 1,048,575 nodes. Facets were extracted from the standard solid, resulting in a total of 21 facets. The NodeIDs were then associated with the facets based on the 3D coordinates.

[0025] Table 1 Node Data Example Step S20: Forward Molding Simulation and Process Field Data Acquisition This case study focuses on the simulation of injection molding of optical lenses, with polycarbonate as the selected material.

[0026] Based on the above-mentioned optical lens physical mesh, the product flow channel, fan-shaped gate, cooling water channel, etc. are further set. The upper surface of the lens is set as the light-incident surface S1, and the lower surface is set as the light-outceasing surface S2. In addition, in order to reduce the residual stress gradient of the optical lens, an overflow area is added. The product simulation model is shown in Figure 3.

[0027] The molding mode is machine mode, and the preliminary process parameters are set as shown in Table 2. In order to ensure that the gate freezing time can be obtained in subsequent processes, the holding time is extended to 30s.

[0028] Table 2 Initial process parameters After completing the injection molding simulation, the nodal point cloud of the standard geometric surface is first extracted based on the mold flow mesh, denoted as: X std,i = (x i ,y i ,zi ); and output the warping displacement vector of the corresponding node: u i = (dx i ,dy i ,dz i ) The overall warping distribution of the lens is shown in Figure 4.

[0029] Step S30: Calculation of non-uniform compensation amount based on material PVT To calculate the effective scaling factor at the node scale, it is necessary to obtain the temperature and pressure state quantities of the same node at the "end of pressure holding time (hereinafter referred to as EOP time)" and the "room time" (hereinafter referred to as room time), i.e. This study defines room temperature conditions as... Pressure conditions are defined as follows: The EOP (Exit Optimization) time is determined by the gate flow rate-time curve: when the gate flow rate drops to a threshold... When the flow rate is below cc / s, pressure transmission is considered to have essentially ended, and this moment is defined as the end of pressure holding. The gate flow rate-time curve is shown in Figure 5, and examples of the output of the corresponding nodes are shown in Table 3.

[0030] Table 3 Example of room time node data The polycarbonate used in this experiment employed an isobaric heating mode for PVT testing to obtain the specific volume-temperature-pressure relationship. The PVT test results are shown in Figure 6. A PVT model suitable for simulation model fitting was obtained using the two-domain Tail equation, with the following formula: in, Specific volume, cc / g; ρ is the specific volume at zero pressure (P=0), cc / g; C is an empirical constant; P is the pressure, MPa; B is the volume modulus, MPa. The parameters of its two-domain Tait equation are shown in the table.

[0031] Table 4. Parameters of the Two-Domain Tait Equation Based on the above temperature and pressure results, the time of pressure holding termination and the specific volume at room temperature were calculated using the two-domain Tait equation, and the volume ratio was also calculated. and effective amplification factor S eff 。

[0032] Record nodes The volume ratios at EOP and room times are respectively Then the volume ratio of node i at time EOP to time room for Effective magnification factor of node i for Calculated volume ratio and effective amplification factor As shown in Table 5: Table 5. Volume ratio and effective magnification factor Step S40: Non-uniform compensation point cloud generation and surface reconstruction preparation To ensure that the compensation direction is consistent with the warping mechanism, this paper adopts a deformation vector-driven compensation method. The nodal warping displacements obtained from forward simulation are then used. As a representation of the offset of this node from standard geometry to final-state geometry, it contains both direction and magnitude information. The goal of compensation is to introduce a reverse perturbation on the mold side to counteract this offset; therefore, the compensated cavity target point cloud is defined as... in Decide on the direction of compensation. Adjust the compensation amplitude. This method can directly obtain the cavity reverse compensation coordinates at the node level, transforming the compensation process from empirical scaling to a quantitative model based on material PVT and warpage prediction.

[0033] Taking NodeID 156063 as an example, according to ,as well as The effective magnification factor was calculated. Calculated Similarly, the compensation results for other example nodes can be obtained, as shown in Table 6.

[0034] Table 6 Example Node Compensation Calculation Results Step S50: Constrained surface fitting and cavity solid generation Using the original product model as the geometric reference, stable features and boundary information are extracted from the original model and mapped onto the compensated point cloud to obtain boundary constraints and topological framework. Based on this, high-precision surface fitting is performed on the optical working surface, and continuous splicing and combination reconstruction between multiple surfaces are completed to finally generate the inversely compensated solid model. The compensated solid model reconstructed based on the point cloud data obtained in step S40 is shown in Figure 7.

[0035] To evaluate the effectiveness of the proposed point cloud inverse compensation method based on real PVT in improving the forming accuracy of optical lenses, this case study employs three schemes for comparative analysis, comparing both surface accuracy and optical performance. The three schemes are defined as follows: Scheme (A) involves direct forming of the standard model without any geometric compensation; Scheme (B) involves overall uniform compensation, referencing a PC material shrinkage rate of 0.6%, and enlarging the mold cavity by a factor of 1.006; Scheme (C) involves non-uniform compensation, i.e., the method proposed in this invention. The surface PV and RMS of the three schemes are shown in Table 7.

[0036] Table 7. PV and RMS of the three molding schemes Compared to scheme (A), scheme (C) shows a significant improvement in surface PV and RMS, and compared to the traditional overall uniform compensation scheme (B), the improvements are 25.26% and 44.56%, respectively. The surface profiles of the three schemes after molding are shown in Figure 7. Meanwhile, the optical performance (birefringence, optical path difference, modulation transfer function MTF) of the three schemes after optical analysis are shown in Table 8.

[0037] Table 8 Optical performance of the three schemes after molding The results of birefringence, optical path difference, and MTF show that the optical performance of scheme (C) is superior to that of schemes (A) and (B), demonstrating the advantages of this method.

Claims

1. A method for reverse compensation of the cavity in an injection mold for optical-grade parts, characterized in that... Includes the following steps: Step S10: standard surface shape definition and standard point cloud discrete modeling, obtaining a standard model of an optical grade injection molded part, determining an optical function aperture area; discretely sampling the standard surface shape model to generate a simulation grid, outputting standard point cloud data X with unique node serial numbers and three-dimensional coordinate information std，i (x, y, z); extracting feature surfaces and surface domain boundary information from the standard model; Using the standard point cloud 3D coordinates as a reference, the standard point cloud data node sequence number is associated with the surface region information and feature boundary; Step S20: Molding simulation and process data acquisition. Set molding process parameters, including but not limited to mold temperature, plasticizing temperature, injection speed, holding pressure, holding time, and cooling time; execute molding simulation under the molding process parameters, and output node deformation vector data u consistent with the node number. i =(dx i ,dy i ,dz i ); Step S30: Based on the non-uniform compensation calculation of the material PVT, determine the reference time as the end time of pressure holding or the end time of pressure transmission, and determine the target state as room temperature and atmospheric pressure; extract the temperature and pressure of each node at the reference time, and obtain the specific volume v of each node at the reference time and the target state in combination with the material PVT model. EOP,i v room,i Calculate the node volume ratio λ; calculate the equivalent magnification factor S of each node based on the node volume ratio. eff Using the node deformation vector as the compensation direction and scaling the compensation amount according to the equivalent amplification factor, a node-scaled compensation vector is constructed. Step S40: Non-uniform compensation point cloud generation and surface reconstruction preparation. Using the node index as a reference, the scaled compensation vector is applied to the standard point cloud to generate non-uniform compensation point cloud data X. cav，i The non-uniform compensation point cloud is used to characterize the geometric morphology of the compensation cavity; The non-uniform compensation point cloud is partitioned into surface regions based on the node sequence number, and the closed boundary point set of each surface region is extracted for subsequent surface fitting, trimming and splicing constraints. Step S50: Constrained surface fitting and cavity solid generation. Under the constraint of closed feature boundary, the point set of each surface region is fitted to generate surface patches. Positional continuity and tangential continuity constraints are applied at the splicing of adjacent surface patches. The surface patches are then stitched and solidified to generate a compensated cavity solid model.

2. The method for reverse compensation of the cavity of an injection mold for optical-grade parts according to claim 1, characterized in that: The material PVT model is obtained based on the isobaric heating path and is used to characterize the specific volume change relationship of the material under heating-pressurization conditions; the nodal volume ratio is determined by the specific volume ratio between the reference time and the target state.

3. The method for reverse compensation of the cavity of an injection mold for optical-grade parts according to claim 1, characterized in that: The non-uniform compensation point cloud mentioned in step S40 is obtained by applying a scaling compensation vector to each node of the standard point cloud based on the node number. The scaling compensation vector is the product of the equivalent magnification coefficient and the node deformation vector.

4. The method for reverse compensation of the cavity of an injection mold for optical-grade parts according to claim 1, characterized in that: The reference time is set as the time when the pressure holding period ends.

5. The method for reverse compensation of the cavity of an injection mold for optical-grade parts according to claim 1, characterized in that: The standard point cloud data, node deformation vector data, and time-series data of node temperature and node pressure are all associated and read based on a unified node sequence number, realizing the calculation of compensation amount and generation of compensation point cloud corresponding to each node.

Images