沖壓模具外文翻譯-汽車后底板的沖壓模具設計分析【中文4170字】【PDF+中文WORD】,中文4170字,PDF+中文WORD,沖壓,模具,外文,翻譯,汽車,底板,模具設計,分析,中文,4170,PDF,WORD ELSEVIER  Journal of Materials Processing Technology 55 (1995) 408-416  Journal or Materials Processing Technology Analysis of die designs for the stamping of an automobile rear floor panel Fuh-K uo Chen Jia-Hong Liu Department of Mechanical Engineering, National Taiwan Universi吵，Taip凹，Taiwan, ROC Rec在ived J O October 1994 Industrial summary The stamping process of manufacturing a one-piece rear floor panel of a passenger car has been investigated in the present study. An analysis of the original die design, in which a split defect occurred at the drawn-cup wall, was performed using circle-grid analysis as well as the 3-D finite-element method. The split defect is due to the large area of sheet metal under the blank-holder that limits the flow of the metal towards the cup area. An optimum die design, which consists of a separate die face and a wedge mechanism mounted in the lower die frame, was proposed to provide additional metal for the cup area and to eliminate the split defect without adding an extra operation. This optimum die design was validated by the results of circle-grid analysis performed for the first and second draw operations and also by the good-quality panels produced. Keywords: Stamping dies; Rear floor panel; Splits; Circle-grid analysis. 1. Introduction Splits are amongst the major defects occurring com- monly in the stamping process. A lot of research effort has been made to investigate the causes of and solutions to the split problem during the last few decades [1-4], the methods used including forming-limit analysis and the finite-element method. Since Keeler and Backofen [5] first introduced the concept of forming-limit diagrams (FLDs) in 1963, they have been used widely in the ana- lysis of sheet metal forming in press shops. The FLDs indicate the strains which lead to failure and thus provide a useful tool to determine if the forming process is likely to be prone to splitting, whilst the finite-element method can calculate the strain distributions in the stamped parts accurately and thus predict if the split defect is likely to occur. In general, the solution to problems of splitting is to provide more metal to the critical area before the major drawing process starts. This can be achieved either by decreasing the blank-holder pressure or improving the lubrication conditions, but the most straight-forward method is to add an extra operation solely for the pur- pose of feeding more metal into the critical area. * Corresponding author. 0924-0136/96/$15.00 ⑥ 1996 Elsevier Science S.A. All rights reserved SSDI 0 92 4 -0 13 6 ( 9 5 ) 0 2 0 3 8 - N However, the extra operation increases the production cost by adding one more set of dies and additional man-power; and hence in reality, it should be avoided. In the present study, an optimum die design in which a separate die face and a wedge mechanism mounted in the lower die frame is proposed to eliminate the occur- rence of a split-defect in the stamping process for a one- piece floor panel of a passenger car. The special die face and wedge mechanism were designed to provide addi- tional metal for the critical area where the split defect occurred, without adding an extra operation. Both circle- grid analysis and 3-D finite-element simulations were performed to analyze this split defect. 2. Problem description The common design for a rear floor panel of a passen- ger car is usually a two-piece type, namely, welding two stamped pieces together, as shown in Fig. 1. The two- piece type design is chosen mainly due to the difficulty encountered in stamping a one-piece rear floor panel, in which a split tends to take place at the wall of the deeply drawn cup used for storing the spare tire, as shown in Fig. 2. The occurrence of the split is attributed to the substantial distance between the cup wall and one side of the blank-holder, as shown by line A-B in Fig. 3, that F.-K. Ch凹，J.-H. Liu / Journal of Materials Processing Technology 55 ( 1995) 408 41幣 409 spl i t  Fig. I. A two-piece rear floor panel. Fig. 2. A split occurring at the cup wall.  The original procedure in the press shop for the pro- duction of the one-piece rear floor panel consists of four operations: first draw, second draw, trimming, and flang- ing. The first-draw operation was designed only to pro- duce a cup shape, as shown in Fig. 3. As for the ribs around the cup, these were formed in the second-draw operation. Like most stamping processes, the main defor- mation of the rear floor panel is completed in the first- draw operation. The conventional draw process allows the punch to pull more metal into the die cavity from the blank-holder. To facilitate metal-flow, no draw beads were employed on the blank-holder surface. H owever噸 due to the large draw depth and the geometric difficulty mentioned above, a split was still found at the cup wall near the bottom after the first-draw operation, as shown in Fig. 2. The location of the split defect indicates that the considerable distance between one side of t he cup wall and the blank-holder does prevent the metal from flow- ing towards the cup. Some efforts have be巳n made to help the metal to flow toward the cup area. The attempt of decreasing the blank-holder pressure led to more wrinkles at the root of the cup area but without eliminat- ing the split. Improvement of the sheet metal quality was also proven to be in vain. Extreme care taken with the lubrication conditions can ease the split problem: how- ever、this is not cost-effective for the process of mass production. Also the large quantity of lubrication oil used in the stamping process may pollute the shop floor. Hence, the most efficient way left to solve this problem is to provide more metal for the cup area before the punch starts to form the cup. To achieve this end, modification of the blank-holder surface to form a better binder-wrap shape which provides more metal around t he cup area was considered. The binder-wrap is the deformed shape of the sheet-blank at the closure of the blank-holders. However, due to the same geometrical reason, i.e. the considerable distance between the cup and one side of the blank-holder, the optimum blank 占older surface is not 巳asy to obtain. Finally, a separate die face designed for the first-draw operation aided by a special wedge mechanism mounted in the lower die fram巳 provided more metal for the cup area and enabled the production of sound prod- ucts without split defects. B Fig. 3. The location of the drawn cup in the sheet-blank. restricts the metal under the blank-holder from flowing into the cup area. Whilst in the two-piece design this distance is much shorter and sufficient metal can flow easily into the cup to prevent the edge of the cup from splitting, due to cost-effective considerations, a one-piece rear floor panel is always desired, so that the split prob- lem must therefore be overcome. 3. Analysis of the original design The problem of splitting is usually related to the strain distribution in the critical area. The strain distribution in any cross section of the formed part is determined by two factors: one is the amount of metal-flow resulting from draw-in over the blank-holder the other is the amount of stretch created by the contacts between the punch and the die [6]. To quantify the effect of the tooling geometry on the metal flow, the original design was analyzed by circle-grid analysis (CGA) and the finite-element method (FEM). 410 f飛 K. Chen, J.-H. Liu / Journal of Materials Processing Technology 55 (199刀 408-416 。.3←一 ，·：／二－ 十 － 0.2 Jf － 0.1←… 。 。 。 .1  。.2 0.3 TRUE STRAIN  。.4 ’，5 Fig. 4. The stress-strain curve for the sheet-blank. 3.1. Circle-grid αnαlysis Circle grid analysis has been used widely in the press shop to measure the strain distributions and thus to enable an analysis of the formability of the sheet metal by plotting the measured strains on the forming-limit dia- gram. The circle grids have a major advantage over the other types of grid, such as square grids, since they do not have any preferred orientations. This advantage lies in that the deformation of the circles will result in ellipses, the two principal directions being displayed clearly by the major and minor axes. The magnitudes of the princi- pal strains can then be calculated from the measurements of the lengths of the major and minor axes. By plotting these measured major and minor strains on the forming limit diagram, the severity of areas of a formed part can be evaluated. In the present study, a production floor-panel made using the original die design was first analyzed with circle-grid analysis. The steel sheet used for production is 0.7 mm thick and of DDQ quality, with material proper- ties as shown in Fig. 4. The corresponding forming-limit diagram for this material provided by the steel supplier is shown by the solid curve in Fig. 5. Any area of a formed part which is deformed with the strains located on or close to this curve will tend to split. In practice, the forming-limit curve shifted down by 10%, as shown by the dashed line in Fig. 5, is used as the design curve. The area above the forming limit curve is called the failure zone; the area between the forming-limit curve and the design curve is termed the marginal zone; and the area below the design curve is named the safe zone. In general, the strain distributions in any area of a formed part should fall in the safe zone to make the stamping process stable, a stamping process being said to be stable if it is less sensitive to process variations. Before being stamped, the critical area of the sheet- blank was imprinted with 5-mm diameter circles with 6-mm spacing between their centres. To imprint the circles the critical area of the sheet-blank was first cleaned using a special cleaner, then a stencil with the correct grid pattern was placed in position on the part. Using the electrolyte as conductor, the area covered by the stencil was marked with the grid pattern by an etching process. To prevent the marked area from rust- ing, a piece of cloth saturated with cleaner was used to wipe off excess electrolyte and residual oxides in the marks. After being stamped, a split was found on the cup wall near to the top, as shown in Fig. 2. The major and minor strains of the deformed circles around the split, as shown in Fig. 6, were measured and plotted on the forming limit diagram, as shown in Fig. 7. As can be seen in this figure, the measured strains are either above or close to the nu r3 F.-K. Chen, J.-H. Liu / Journal of Materials Processing Technology 55 (1995) 408-41幣 411 Mα-ginal ...... 4” 氣afe z 附 《U F3 n3 au oO吁， nu r3 nu r3 nU FD nu rD nu n3 06 00 句’ ”’ FO rb r3 70 「 \ 65 占。60 ..._., 55 E3 TF） 50』舊 的 C 擊 45 ＼ ＼ ＼ ＼ ，， --- r3 nU ’句 ·.0.. so ，， ＼ ‘產 ／ ’句 句3 句a q‘， 』 。「OE 』 40 E3 nu rD AU r3 nu r3 旦 35 EJ nU FD nu 句’＆ 句’－ ‘1 句4 41 嚕l 豆 30 氣、/ c3 4l -30-25-20-15-10 -S O 5 10 15 20 25 30 35 40 M i no r Strai n （句 ） Fig. 5. The forming-limit diagram for 0.7 mm DDQ steel sheet. Fig. 6. Deformed circles on the sheet-blank. forming-limit curve, and the failure is obviously due to stretching, since both the major and minor strains are positive. It is also noted that the strains are very close to the plane-strain failure mode, i.e., close to the axis on which the minor strain is zero. The results of the circle-grid analysis indicate that the original design is very unstable. The FLD also indicates that the major strain is too large: this is consistent with the present authors' opinion that the considerable dis- tance between the cup and one side of the blank-holder limits the flow of metal towards the cup area, resulting in the large strains. As discussed in the previous sections, -30-25-20-15升。5 0 5 10 15 20 25 30 35 40 M i n or Stra in (°lo) Fig. 7. Measured strains around the split. the most e仔ective method of decreasing the major strain is to provide more metal for the cup area. 3.2. Finite element anaysis To help further understanding of the deformation of the sheet-blank during the stamping process, a 3-D finite- element analysis was performed for the first-draw opera- tion of the original design. The explicit finite-element code PAM-STAMP, which is capable of handling any arbitrary 3-D die shapes, was used to conduct the simula- tions. Since the 3-D die geometries, including the punch, the die cavity, the blank-holder and the draw beads, are not simplified, the finite-element program is able to simu- late the actual production processes more accurately. In order to describe the geometry of the die compo- nents, a commercial CAD program was used to construct the surface models for these components. The mesh sys- tems required by PAM-STAMP as the input data for the die geometries were then generated by the same commer- cial CAD program, as shown in Fig. 8. In earlier periods, it was very difficult, if not impossible, to generate the mesh system for a complicate 3-0 die shape, such as the stamping dies. However, as CAD systems are being used with increasing popularity in the die and mould industry, the above procedures to generate the mesh systems for the die geometries become trivial. Since the PAM- STAMP code treats the die components as rigid bodies, the mesh systems are used only to describe the geometries 412 F.-K. Chen, J. H. Liu / Journal of Mα，terials Processing Technology 55 (1995) 408→416 blank-holder d ie sheet-blank  Fig. 8. Mesh system for the original die design. pu nch of these components and not for stress analysis. In the present investigation, a mixture of 3-node triangular and 4-node rectangular elements is used to construct the mesh systems. The mesh system for the sheet-blank is also shown in Fig. 8. As seen in this figure, the mesh density is much higher in the cup area than elsewhere, since the cup area is the location where splitting occurs. The numbers of elements used in the analysis are sum- marized as follows: die: 9,910; punch: 5,499; blank-holder: 4,411; sheet: 4,891; total: 24,711. The material properties used for the finite-element analysis are the same as those given in the previous section, the other operational conditions being: blank- holder pressure: 57 kPa; punch velocity: 10 m/s; punch stroke: 895 mm; and coefficient of friction: 0.12. The CPU time spent on an HP735 work-station to run a single job is about 11 100 s, the simulation results being displayed on an SGI work-station. One of the advantages of using finite-element analysis for stamping process is that the deformed shape of the sheet-blank can be monitored throughout the stamping process, which is not possible in the real production process. Amongst the deformed shapes, the binder-wrap is the most interesting to the die designer, since it can be used to determine if the blank-holder surface has been designed properly. In the present study, the main purpose of the finite-element analysis is to investigate the defor- mation of the sheet-blank and the location where the split is most likely to occur. Hence, the deformed shape and the thickness contour obtained from the finite-ele- ment simulations were investigated. The deformed mesh is shown in Fig. 9. As seen in this figure, the cup is formed without causing any wrinkling. As for the thickness contour shown in Fig. 10, the darker area represented the thinner portion of the steel sheet. It can be seen from this figure that the thinnest portion is at the cup wall near to the bottom: this location agrees with that found in the production panel. In addition, the distribution of the major and minor principal strains around the thinnest portion obtained from the finite-element simulation, as shown in Fig. 11, is very consistent with that found in the real production panel, as shown in Fig. 7. Hence, the finite-element simulation confirms the previous circle-grid analysis and is validated by the production process. With the validated process parameters, the finite-element analysis F.-K. Chen, J.-H. Liu / Journal of Materials Processing Technology 55 (1995) 408 41幣 413 Fig. 9. Deformed shape for the original die design. can be used for the analysis of any modified die design to substitute for the real die try-out. Compar巳d with the real die try-out, the finite-element simulation is not only cost-e仔ective but also time-saving, the example THICKNESS …CONT：OUR VALUES 5.209￡-01 5.718￡-01 6.227￡-01 ...= 6.737￡-01 c:m = 7.246￡-01 區(qū)軍丑 7.756￡-01 c==i :8.265￡-01 MIN = 3.680￡-01 IN ELEM. 10207 MAX = 8.775￡..()J IN ELEM. 10319  given in the following section demonstrating this advant- age. 4. A modified die design Since the split-defect is due to the lack of metal around the cup, one possible modification is to open the blank-holder at one side of the cup, as. shown in Fig. 12, which may cause more metal to flow into the cup. The geometries of the die and the sheet-blank remain the same as in the original design and同 hence, are not shown in Fig. 12. In order to validate this modification, a 3-D finite 巳lement simulation was performed instead of re-vamping t he real die. The simulation conditions were the same as those for the original design, except that the geometries of the punch and the blank-holder were changed 、as shown in Fig. 12. The distribution of the major and minor strains of the whole panel obtained from the finite-element simulation …：4.699￡-01 …： …：：4.1仰 01 Fig. I 0. Thickness contour obtained from 白nite-element simulation for the original die design 414 F.-K. Chen, J.-H. Liu / Journal of Materials Processing Technology 55 (1995) 408 416 ＼ ＼ ＼ 三 60 ＼ 飛 C 、、 .a.. ＼ U飛 ＼ ，， ＼ ＼ ＼ 。 寫 60 ＼ c－HU』 ＼ ＼ - ＼ ， ＼ 咱回， ＼ ，二 奇J ＼ 。。40 ＼ 」 二 －－－ ＼ － ， (/) 40 乏 ，， ， 」O「OF之 、“ ”’ ＋ ＋ 4← 20 -30 -20 -10 0 1.0 20 30 M i nor Strain (0／，） Fig. 11. The strain distribution obtained from finite-element simulation for the original die design . 20 -30 -20 -10 0 10 20 30 M i nor Stra i n (°lo) Fig. 13. The strain distribution for the modified die design. ν ?鎖在：ti跟你 ι ． ‘．？： ．．．．．‘ ．、豐．】一 －．♂ ．．．．．電‘、‘：‘...，‘ ．．‘‘．‘ 啕 附則4 ． ‘ ．‘ ， 、“三“’ v 一·“？‘ 一’ punch blank-holder Fig. 12. The modified die design.  Fig. 14. The deformed shape for the modi且ed die design. for the modified die design is shown in Fig. 13. As seen in this figure, due to a lot of material being drawn into the cup from the unconstrained area, the strain distribution moves down a little bit but is still in the marginal area. Fig. 14 shows the deformed shape, in which serious wrinkles are observed in the unconstrained area. Al- though the split problem might be avoided in this modi- fled design, the serious wrinkles are unacc將ptable. As a result, the modified design was not feasible according to the 3-D finite-element simulation, and the re-vamping of the real die was avoided. F.-K. Chen, J.-H. Liu / Journal of Materials Proce.,sing Technology 55 (1995) 408 41幣 415 5. The optimum die design As discussed in the previous se