Text/Tao Haisheng DPCA Automobile Co., Ltd
In view of the problem of peugeot 2008 front fender A-side rebound ultra-poor, through testing, verification, and finally converted into process digital modules for precise processing, the problem was solved at one time. The solution to the problem of rebound and ultra-difference has a certain representative effect on the solution of the problem of auto body appearance parts, especially the A-side rebound problem, has strong implementability and generalizability, and has a certain reference role in how to improve the appearance quality and dimensional accuracy of body products.
The development speed of the domestic automobile industry is unparalleled in the world, and the global brands are gathered together, and the competition is becoming increasingly fierce. A car with a high appearance, in addition to the good design, the workmanship must also be good. Fine, stable, good sealing (sound insulation and waterproofing), these are inseparable from the improvement of the geometric accuracy of the body stamping parts. Among them, the gap surface difference is the quality characteristic that ordinary consumers can perceive. The general dimensional difference is good to rectify, but it is difficult for the appearance parts, especially the rebound control of the appearance parts. In most cases, if the appearance of the stamping cover mold is completed, if the problem of rebound and size difference that remains is not very serious, it can be solved by driving the outer plate by the inner plate and adjusting the size of the fixture; if it is not easy to be perceived and difficult to solve, it is generally abandoned at the production stage.
Near the start of the Peugeot 2008 project, there was a significant surface difference between the overlapping part of the front fender and the hood, which was checked and determined to be caused by the rebound of the front fender arch. Considering that the front fender is an exterior part, the A-side area, can not easily change the profile, dragged for a long time the problem of surface difference is still difficult to solve, resulting in the vehicle appearance quality inspection has not been able to pass, so it is necessary to find the fastest and most effective way to solve the problem in the shortest time.
Introduction to the process plan and description of the problem
Introduction to the process concept
The Peugeot 2008 front fender machining process scheme is shown in Figure 1 (symmetrical side omitted).
Figure 1 Peugeot 2008 front fender process scheme
Problem description
Peugeot 2008 front fenders and the hood overlap part has a section of arching, as shown in Figure 2, described in stamping terminology, that is, rebound caused by dimensional overshoot, poor problems behind the assembly affect the appearance of the vehicle. The 3D detection position of the fender is shown in Figure 3, and the detection results are shown in Table 1, and the red part of the table is the variance term. The difference item in the appearance of A side, with ordinary product modeling compensation to correct, there is a greater risk, because the rebound operation and compensation is difficult to do 100% accurate, and there is a risk of destroying the A side modeling, if not done well will cause greater appearance defects.
Fig. 2 The overlap surface of the fender and the hood is different
Figure 3 7 3D detection positions for the fender
Table 1 Test results
Problem cause analysis
Analysis of the difference between samples in each process
Under normal circumstances, starting from the test results of the product, it is necessary to figure out where the product's variance is, which can be directly seen in the daily 3D test report and the feedback of the quality department. From the 3D inspection results, the rebound deviation area of the part has been found, so the analysis of the process sample can be detected from the OP40. Because the process sample is different from the final product, many parts are not formed in place and cannot be clamped on the detection bracket for inspection, so a more convenient blue light scan detection is selected. The OP40 blue scan results are shown in Figure 4, and the yellow areas in the figure indicate the rebound overdression areas; the OP30 blue light scan results are shown in Figure 5, and all green indicates that there is no rebound.
Fig. 4 OP40 blue light scan results (yellow is the rebound deviation part)
Fig. 5 OP30 Blu-ray scan results (all green without rebound)
Cause analysis of the rebound
By analyzing the sample size of each process, it was found that negative rebound was generated after the OP40 flanging process. Experience tells us that the main reason for negative rebound is that the material is not unfolded for a long time after flanging, and the internal stress release occurs shrinkage deformation to produce negative rebound. Objectively, the OP40 will be stretched and deformed when the unfolding length becomes longer after the flanging, but carefully comparing the shape changes before and after the flanging, it is believed that it may be a negative rebound at the 2 bosses after the flanging due to the need for material filling, which is only an analysis judgment, and it is necessary to verify and find the exact value of the rebound compensation. Because of the time relationship, if the CAE forming simulation analysis is used, the time cycle and manpower investment are relatively large problems, and generally can only achieve 80% to 90% accuracy, so a convenient and reliable physical test verification method is selected. If the physical product will be subject to a large cost through the NUMERICAL control test, it is selected to do manual samples for testing first. Find a plate of material, knock out 2 small steps on the sleeper, and press out the hand sample as shown in Figure 6. The 3D detection found that the rebound amplitude at the highest point was reduced by 0.2 mm, which basically determined the factors affecting the rebound and the direction of solving the rebound.
Fig. 6 First manual sample (directionality test)
Verify the solution
In order to further verify the feasibility of the scheme, it was decided to make three new manual samples by hand, this time on the jaw table, and the shape of the 2 bosses was greatly improved with a small hammer, as shown in Figure 7. The factors to be considered when making samples are: the post-process flanging deformation, the tensile deformation of the material; the hardening of the material, the size can not be small; the shape of the parts pressed out of the mold is more uniform; the fillets should be enlarged so that there will be no defects in the secondary forming of the rounded corners when the subsequent flanging; whether the trimming process can successfully complete the trimming; the trend of the quality and size of the forming appearance after the post-process flanging is estimated.
Fig. 7 Second manual sample (precision test)
The three handmade samples that were made were taken to the post-process mold and formed to obtain a relatively ideal appearance quality, as shown in Figure 8, the geometric dimensions were qualified, and the arch rebound of the product was basically eliminated. This proves that the analysis of the problem is correct and the solution is feasible.
Fig. 8 Parts pressed out of the second manual sample
Mold adjustment
The appearance of the mold change, you need to judge the risk and feasibility. When the mold dealer hears our suggestions for adjusting the mold, it is difficult to accept, because the mold adjustment workflow is complex, including complex processes such as process digital mold change, form analysis, rebound compensation, expert review, etc., and the assembly line operation requires a relatively long cycle. After repeated communication between the two sides, the mold manufacturer accepted our suggestion to change the process digital mold and completed the rectification of the mold. The changed process digital module is shown in Figure 9, the op20 sample effect after the mold change is shown in Figure 10, and the simulated pressure test part effect is shown in Figure 11, and the effect is perfect.
Figure 9 Process digital modulotype change
Figure 10 Op20 sample effect after mold change
Fig. 11 Simulate the effect of the test pressure part after the mold change
Summarize and analyze and promote applications
Advantages of physical verification
According to the author's experience, forming analysis is the preferred solution to the problem, but sometimes it is limited by some objective conditions, and it is more necessary to consider usability and efficiency. Physical verification and simulation calculation, the former is 100% reliable, does not require cost input, and the industry for CAE accuracy evaluation of the best about 90%, may need to do a round or two of rectification. Obviously, this hand-made verification method is simple, practical, flexible, fast and reliable.
The simulation verification of the whole process before and after the mold change is compared with the actual product
Because of the urgency of production, the most convenient method was used to solve the problem, and then CAE was used for verification and summary. We replaced the original process digital mold with the changed process digital module, compared the whole process simulation of the two states, and found that the modified process digital mode was formed and analyzed, and the simulation results obtained were basically consistent with the results of the manual sample test described earlier. In the simulation of the whole process before compensation, the maximum rebound exceeded 1.3mm. Seemingly simple work, including a lot of thinking, calculation, experience reference and manual correction, such as the shape of the final process of the digital module and the shape of the manual sample have been corrected, the main consideration of the sample is local forming, hardening is more serious, for the mold after the change of the forming, is the uniform forming of the entire sheet, for this reason in the process of digital mold, the boss size is appropriately reduced. The simulation results of the intermediate process before and after the change of the process digital module are shown in Figures 12 and 13, and the final product simulation results are shown in Figure 14, and the simulation results show that the final product rebound amount (the gap between the theoretical position of the measurement point) before the process digital module is changed is a large area of the deviation (deviation value in the red line area), and the final product rebound amount after the process digital module is changed is all within ±0.5mm, and no rebound error is found.
Figure 12 Comparison of simulation results before and after the OP20 process digital analog change
Figure 13 Comparison of simulation results before and after the OP30 process digital module change
Figure 14 Comparison of the final product simulation results before and after the process digital module change
The difference between manual samples and process digital modules
As mentioned earlier, when changing the process digital module after the manual sample verification, the boss size was reduced, because the manual sample was hardened, the mold sample was formed more uniformly, and the final result proved that this analysis judgment was accurate. The reduction of the boss size before and after the final product and the simulation results are shown in Figure 15.
Figure 15 Comparison of the final product with the simulation results before and after the reduction of the boss size
Verification of the accuracy of the forming simulation
With the accuracy of CAE analog forming analysis, it is better to achieve 90% accuracy. The old process digital analog simulation results show a maximum rebound value of 1.3mm, and the maximum rebound value of physical detection is 1.06mm (blue light scanning results, 3D detection results are clamped state, not authentic), can be estimated as 1mm, obviously the simulation shows a rebound value larger than the real thing. The physical detection value is small, or it may be that the mold has been rebounded before delivery, reducing the rebound amplitude, but it has not been completely solved.
The advantage of forming analysis is that it can be simulated by computer operation, there can be more choices for testing, and it is easy to achieve directivity judgment for rebound. As for the calculation accuracy, the author agrees with the consensus in the industry, for the steel plate below H440, including the general plate, if the design of the process digital module plus practical experience and optimization, the simulation result and physical consistency are about 90%, which is an ideal result.
Application
The methods and processes discussed in this article summarize the problems, which have certain promotion and application value, for all appearance parts, especially the opening parts, in addition to the A-side compensation, which is a relatively stupid method, it is necessary to find more ways to supplement the process, solve this problem between processes, and solve the problem more easily and reliably through manual or CAE calculation compensation.
Springback is a common phenomenon, most of which is solved by following the inner plate according to the outer plate, or by clamping the welding assembly and the adjustment of the whole vehicle during welding. If the rebound problem is not solved, there will be certain defects in the appearance of the whole vehicle, and over time, it will also cause deformation due to the release of internal stress. The well-sized stamping parts not only bring sensory effects to the production of the whole vehicle, but also have dimensional stability, which has obvious effects on the sealing, noise reduction and water leakage prevention of the whole vehicle.
——Article from: Forging and Stamping, No. 4, 2022