Reverse modeling and die-casting mold design of aluminum alloy die-casting part
Time:2024-10-21 08:48:33 / Popularity: / Source:
Abstract
In order to quickly respond to market and improve R&D efficiency, shell of an aluminum alloy signal amplifier is reverse modeled and die-casting molds are designed. First, ATOS blue laser scanning system is used for measurement, then Geomagic Design X software is used to process and reverse model measured point cloud data; after die-casting process analysis is carried out according to reverse prototype, die-casting mold design of product is combined with die-casting mold design theory and experience. Research process shows that reverse modeling method based on laser scanning and Geomagic is fast and accurate in size. Through product reverse design and empirical mold design methods, new products and their die-casting molds can be quickly developed.
Die casting is an efficient near-net forming process, and die castings have been increasingly used in various industries in recent years. With intensification of market competition, high efficiency, low cost, personalization and innovation have become core of enterprises to improve their competitiveness. In the field of product design, traditional forward design and measurement methods have been difficult to meet development requirements of "fast, good and economical" in many cases, which has led to rapid development and application of reverse engineering (RE) technology. RE can convert real three-dimensional information into computer-processable point cloud data through different measuring equipment, so that processing results can be used for computer-aided communication (CAX) in different occasions to improve design and production efficiency.
This paper uses ATOS system based on blue light technology to reverse measure and model a certain aluminum alloy signal amplifier shell part, then imports model into 3D drawing software to directly split part, and designs die-casting mold of shell part.
Die casting is an efficient near-net forming process, and die castings have been increasingly used in various industries in recent years. With intensification of market competition, high efficiency, low cost, personalization and innovation have become core of enterprises to improve their competitiveness. In the field of product design, traditional forward design and measurement methods have been difficult to meet development requirements of "fast, good and economical" in many cases, which has led to rapid development and application of reverse engineering (RE) technology. RE can convert real three-dimensional information into computer-processable point cloud data through different measuring equipment, so that processing results can be used for computer-aided communication (CAX) in different occasions to improve design and production efficiency.
This paper uses ATOS system based on blue light technology to reverse measure and model a certain aluminum alloy signal amplifier shell part, then imports model into 3D drawing software to directly split part, and designs die-casting mold of shell part.
1 Reverse measurement of parts
Figure 1 shows ATOS Compact Scan 5M optical three-dimensional data acquisition scanner used for part measurement.
Figure 1 ATOS Compact Scan 5M three-dimensional optical scanner
ATOS scanner uses principle of optical triangulation and uses two charge-coupled device (CCD) cameras to detect shape of object and synthesize three-dimensional data. It also supports use of probe detection, which can assist in detection of highly reflective surfaces such as mirrors and deep holes that are difficult to capture with CCD cameras. Scanner can reach benchmark quality level in terms of accuracy, sharpness and completeness. Its technical specifications are shown in Table 1.
ATOS scanner uses principle of optical triangulation and uses two charge-coupled device (CCD) cameras to detect shape of object and synthesize three-dimensional data. It also supports use of probe detection, which can assist in detection of highly reflective surfaces such as mirrors and deep holes that are difficult to capture with CCD cameras. Scanner can reach benchmark quality level in terms of accuracy, sharpness and completeness. Its technical specifications are shown in Table 1.
Project | Parameters |
Scanner size/mm | 34x130x230 |
Camera | 5 million pixels x2 |
Dot pitch/mm | 0.017~0.481 |
Scanning range/mm2 | 40~1 200 |
Working distance/mm | 450~1200 |
Working temperature/℃ | 5~40 |
Power supply (AC) | 90~230 |
Table 1 ATOS Compact Scan 5M Technical Specifications
Considering that inner surface of aluminum alloy part has strong reflection, developer is evenly sprayed on its surface to obtain high-precision measurement results. Since ATOS scanning head needs to scan each surface of shell separately and then merge surface point cloud data, it is necessary to arrange reference points in overlapping area of two surfaces where point cloud data is obtained, that is, to attach black circles and white dots to parts to be measured. System software performs registration calculations and coordinate conversion on three-dimensional data of these reference points, finally realizes splicing of multiple point cloud data. Effect of spraying developer on part and affixing reference points is shown in Figures 2a and 2b, and point cloud data obtained by scanning is shown in Figure 2c.
Considering that inner surface of aluminum alloy part has strong reflection, developer is evenly sprayed on its surface to obtain high-precision measurement results. Since ATOS scanning head needs to scan each surface of shell separately and then merge surface point cloud data, it is necessary to arrange reference points in overlapping area of two surfaces where point cloud data is obtained, that is, to attach black circles and white dots to parts to be measured. System software performs registration calculations and coordinate conversion on three-dimensional data of these reference points, finally realizes splicing of multiple point cloud data. Effect of spraying developer on part and affixing reference points is shown in Figures 2a and 2b, and point cloud data obtained by scanning is shown in Figure 2c.
Figure 2 Signal amplifier housing and its point cloud
2 Reverse modeling of parts
2.1 Point cloud data processing
Import original point cloud data obtained by scanning into Gemo, and perform corresponding preprocessing on point cloud data. First, perform "noise elimination" to delete noise points; "sample" to streamline cloud points according to ratio, then "smooth", finally "triangulate point cloud data" and perform repairs and other processing through software's built-in patch wizard. Effect of point cloud preprocessing and final triangular mesh repair is shown in Figure 3.
Figure 3 Point cloud data processing effect
2.2 Model reconstruction
In reverse engineering, model construction is process of reconstructing point cloud data into entities. System divides surface into multiple areas according to curvature and geometric features of point cloud, and distinguishes them with different colors, as shown in Figure 4a. In order to establish an operating plane and facilitate various operations such as translation, rotation, quick view switching and position determination, center of the circle is combined with plane to align with system coordinates. Then use "face sketch" function to obtain part section data, edit and modify sketch, and generate a closed section, as shown in Figure 4b.
Figure 4 Domain division and section sketch
After establishing part sketch, generate entities through commands such as stretching and rotation (Figure 5a), then make detailed trimming of generated entity to obtain a part model consistent with prototype contour, as shown in Figure 5b.
After establishing part sketch, generate entities through commands such as stretching and rotation (Figure 5a), then make detailed trimming of generated entity to obtain a part model consistent with prototype contour, as shown in Figure 5b.
Figure 5 Part reverse effect diagram
To verify accuracy of reverse modeling dimensions, reverse model is compared and analyzed with pre-processed point cloud (with an upper and lower error limit of ±4 mm), as shown in Figure 6. It can be seen from error scalar color display that inner and outer plane errors of reverse shell parts are extremely small. Errors of selected points in the figure are mainly concentrated in the range of ±1×10-2~±1×10-1 mm, and maximum errors are mainly distributed on inner edge of model, with a number of 1 mm.
To verify accuracy of reverse modeling dimensions, reverse model is compared and analyzed with pre-processed point cloud (with an upper and lower error limit of ±4 mm), as shown in Figure 6. It can be seen from error scalar color display that inner and outer plane errors of reverse shell parts are extremely small. Errors of selected points in the figure are mainly concentrated in the range of ±1×10-2~±1×10-1 mm, and maximum errors are mainly distributed on inner edge of model, with a number of 1 mm.
Figure 6 Comparison of reverse model and point cloud data
It should be pointed out here that error between point cloud obtained by laser scanning and reverse model in this study is mainly concentrated in sealing groove near inner edge of part (Figure 6a). This is mainly due to spraying of developer before laser scanning, consideration of "part design intent" and rounding of local dimensions during reverse modeling. Although rounding and redesigning the model size will lead to large local errors between model and point cloud, this also reflects essence of understanding design intent ("soul") of part in reverse engineering rather than design size ("body").
Absolute deviation and standard deviation distribution of reverse model and point cloud are shown in Figure 7. Combined with detailed analysis file of Geomagic, it can be seen that number of points with absolute deviation values within range of -0.75~+0.75 mm accounts for 97.55%, and according to normal distribution results of standard deviation of measurement points in Figure 7b, number of points within 3 standard deviations accounts for as high as 98.74%, which shows that reverse model is highly consistent with laser scanning point cloud.
It should be pointed out here that error between point cloud obtained by laser scanning and reverse model in this study is mainly concentrated in sealing groove near inner edge of part (Figure 6a). This is mainly due to spraying of developer before laser scanning, consideration of "part design intent" and rounding of local dimensions during reverse modeling. Although rounding and redesigning the model size will lead to large local errors between model and point cloud, this also reflects essence of understanding design intent ("soul") of part in reverse engineering rather than design size ("body").
Absolute deviation and standard deviation distribution of reverse model and point cloud are shown in Figure 7. Combined with detailed analysis file of Geomagic, it can be seen that number of points with absolute deviation values within range of -0.75~+0.75 mm accounts for 97.55%, and according to normal distribution results of standard deviation of measurement points in Figure 7b, number of points within 3 standard deviations accounts for as high as 98.74%, which shows that reverse model is highly consistent with laser scanning point cloud.
Figure 7 Deviation between reverse model and point cloud data
To verify effect of reverse model, Anycubic light-curing rapid prototyping machine with a resolution of 0.047 mm was used for 2:1 printing, and effect is shown in Figure 8. From details of parts such as slots, holes, columns, and reinforcement ribs, light-curing model of reverse model fully reflects prototype features, surface is smooth (dense dots on outer surface of rapid prototype in figure are traces left by removing support), transition between inner, outer corners and edges is smooth. Main features such as contour and slot of metal prototype and rapid prototype were measured with a vernier caliper (accuracy 0.02 mm). Comparison found that maximum dimensional error of rapid prototype was within 0.07 mm, and tolerance level was IT9~IT10. This shows that cumulative error after laser scanning, reverse reconstruction of prototype, model scaling printing can be controlled within a reasonable range, model quality based on laser scanning and Gemo reconstruction meets part tolerance requirements.
To verify effect of reverse model, Anycubic light-curing rapid prototyping machine with a resolution of 0.047 mm was used for 2:1 printing, and effect is shown in Figure 8. From details of parts such as slots, holes, columns, and reinforcement ribs, light-curing model of reverse model fully reflects prototype features, surface is smooth (dense dots on outer surface of rapid prototype in figure are traces left by removing support), transition between inner, outer corners and edges is smooth. Main features such as contour and slot of metal prototype and rapid prototype were measured with a vernier caliper (accuracy 0.02 mm). Comparison found that maximum dimensional error of rapid prototype was within 0.07 mm, and tolerance level was IT9~IT10. This shows that cumulative error after laser scanning, reverse reconstruction of prototype, model scaling printing can be controlled within a reasonable range, model quality based on laser scanning and Gemo reconstruction meets part tolerance requirements.
Figure 8 Model printed by light-curing rapid prototyping method
3 Design of die-casting mold
3.1 Analysis of die-casting process of casting
Signal amplifier shell material is YL112, with a size of 150 mm * ±0.3 mm). Part is a regular rectangular slot shell with external protruding threaded holes (later processing) and a mounting shaft, is equipped with heat dissipation fins and grooves for placing sealants on edges. Combined with characteristics of part and according to principle of parting surface selection, parting surface shown in Figure 9a is determined. Since outer surface of part contains many vertical heat dissipation fins, it is not suitable to choose center pouring, so a flat side gate is selected, as shown in Figure 9b.
Figure 9 Determination of parting surface of die casting and runner design
According to maximum cross-section of die casting and projection area of pouring system, combined with recommended value of 40 MPa injection pressure, clamping force is 1 099 kN, and J1113G horizontal cold chamber die casting machine is selected. Then, after calculating actual capacity of pressure chamber, Φ60 mm was selected to meet design requirements.
According to maximum cross-section of die casting and projection area of pouring system, combined with recommended value of 40 MPa injection pressure, clamping force is 1 099 kN, and J1113G horizontal cold chamber die casting machine is selected. Then, after calculating actual capacity of pressure chamber, Φ60 mm was selected to meet design requirements.
3.2 Die Parameter Calculation
Die-casting machine determines thickness range of die-casting die, stroke of die-casting motor seat determines distance required to remove die-casting. According to size of casting and required safety value, minimum distance required for demolding and removing die-casting is 85 mm, which is less than maximum stroke of die-casting motor seat of 350 mm, which meets requirements; depth of inner runner is 1.8 mm, thickness of fan-shaped runner entrance is 10 mm, width is 8.6 mm, and measured opening angle is 60°, which meets requirements. According to part structure, mold is designed with two overflow grooves, and according to volume ratio of overflow groove of casting, overflow groove radius is determined to be 8 mm, overflow port is 12 mm long, and thickness is 0.8 mm.
3.3 Die-casting mold assembly
Die-casting mold of aluminum alloy signal amplifier is shown in Figure 10. Fixed mold 12 is fixed on fixed mold sleeve 13 on one side of gate sleeve 9, and movable mold 14 is fixed on movable mold sleeve 15. When die-casting machine is working, mold is closed, piston presses aluminum liquid to fill cavity, and mold is opened directly after maintaining pressure. When movable mold and fixed mold are separated to a certain distance that can easily remove casting, it stops. Then, ejector mechanism transmits thrust through ejector fixing plate 5 and ejector 16 to eject casting, and finally movable mold 14 is reset by spring reset mechanism 7.
1. Countersunk hole 2. Moving mold base plate 3. Screw 4. Push plate 5. Push rod fixing plate 6. Pad 7. Reset spring 8. Sleeve plate screw 9. Gate sleeve 10. Guide column 11. Guide sleeve 12. Fixed mold 13. Fixed mold sleeve 14. Moving mold 15. Moving mold sleeve 16. Push rod 17. Limit pin
Figure 10 Mold assembly drawing
Figure 10 Mold assembly drawing
4 Conclusion
(1) ATOS blue optical scanner can accurately obtain metal part point cloud data by spraying and pasting reference points, providing high-precision geometric parameters for reverse modeling.
(2) Based on Gemo, point cloud data preprocessing can be quickly realized, modeling can be realized through a series of operations such as point cloud partitioning-positioning-face sketch operation-stretching-rotation-trimming, etc., reverse model meets part tolerance requirements.
(3) Reverse engineering through ATOS scanning and Gemo modeling can provide prototype parameters for product development and mold design, improving R&D efficiency and market response speed.
(2) Based on Gemo, point cloud data preprocessing can be quickly realized, modeling can be realized through a series of operations such as point cloud partitioning-positioning-face sketch operation-stretching-rotation-trimming, etc., reverse model meets part tolerance requirements.
(3) Reverse engineering through ATOS scanning and Gemo modeling can provide prototype parameters for product development and mold design, improving R&D efficiency and market response speed.
Recommended
Related
- Effect of heat treatment on structure and mechanical properties of die-cast AlSi10MnMg shock tower12-26
- Two-color mold design information12-26
- Analysis of exhaust duct deceleration structure of aluminum alloy die-casting parts12-24
- Research on injection mold for thin-walled inner wheel cover of automobile12-24
- Impact of high pressure casting and rheocasting on salt core12-23