Materials Processing
Development of Forming Technology to Reduce Dimensional Scattering of Automotive Parts with Cambers by Using Bauschinger Effect
2021 年 62 巻 12 号 p. 1750-1756
詳細
2021 年 62 巻 12 号 p. 1750-1756
Dimensional scattering is a severe problem in press forming of ultrahigh-strength steels (UHSS), because of material strength scattering in mass production. In this study, camber back, which occurs in longitudinally curved parts, was examined, and a new forming method whereby dimensional scattering of camber back can be suppressed by the Bauschinger effect was developed. The new method consists of two processes. In the 1st process, a blank is formed with a small radius of curvature compared with that in the 2nd-process. In the 2nd process, that part is formed to a larger radius of curvature than in the 1st process, and the Bauschinger effect is utilized to decrease the amount of camber back. The new method was applied to hat-shaped models of 590, 980, and 1180 MPa-grade steels in which the radius of curvature in the longitudinal direction was 1600 mm. The experimental results showed that the difference in the amount of camber back between the 590 MPa and 1180 MPa steels formed by the developed method decreased by 95% compared with parts formed by the conventional method.
This Paper was Originally Published in Japanese in J. JSTP 60 (2019) 155–160. The abstract is slightly modified.
Developed press forming method using Bauschinger effect.
In recent years, application of ultrahigh-strength steels (UHSS) to automobile parts has been advanced in order to reduce the weight of automobile bodies and improve collision safety.1–4) Most automobile frame parts are manufactured by press forming, which has the advantage of high productivity. However, one problem in press forming of UHSS is poor dimensional accuracy caused by springback.5,6) Because the stress at the bottom dead center of press forming increases as the material strength increases, the dimensional accuracy of UHSS deteriorates due to the increased amount of springback.
Springback includes wall opening due to vertical wall warp and punch shoulder angle change, longitudinal springback in curved parts, and twist in 3-dimensional bent parts. Wall opening is caused by the bending moment due to the stress difference between the surface and back of the blank. On the other hand, springback of curved parts is caused by the bending moment as a structure due to the surface stress at the bottom dead center of press forming. Various countermeasure technologies have been proposed for wall opening in the past.7–9) The proposed measures for springback in curved parts include a method of controlling stresses by changing the width of the product in the pre-process and post-process,10) a method of cancelling the springback factor stress by using a drawing-redrawing process11) and others, but effective countermeasures are still lacking.
In general, the press forming process for automobile parts consists of 3 to 5 processes, and the process for forming the desired shape comprises 2 to 3 steps among them. Conventionally, in forming the desired shape, dimensional accuracy has been maintained by forming the part to a shape close to the product shape in the first process, followed by forming to the estimated shape in the final process.12,13) Highly accurate springback prediction has become possible by the recent development of FEM (Finite Element Method) analysis,14) and the estimated shape can be designed efficiently by FEM. However, since large stress occurs at the bottom dead center of press forming in the final process, when scattering of material strength are large as in mass production processes, the stress fluctuation at the bottom dead center of press forming is also large and dimensional scattering increases.
To resolve this problem, we developed a new press forming method to reduce dimensional scattering of camber back by utilizing the Bauschinger effect.15,16) The developed method is a technology in which the shape in the previous process is designed in such a way that the flow stress reduction effect by the Bauschinger effect is expressed at the bottom dead center of press forming in the final process. By using the developed method, even when forming UHSS having large material strength scattering, the stress fluctuation at the bottom dead center of press forming is reduced, and as a result, the amount of springback and dimensional scattering are also reduced. In this paper, the optimum die shape for the previous process was designed by FEM analysis, and the effect of the developed method was verified by a press experiment.
This chapter explains the mechanism of camber back that occurs in side view curved parts and the aim of the new forming method. Camber back is a type of springback that occurs in longitudinally curved parts, as shown in Fig. 1. This camber back occurs as a result of the bending moment caused by the stress difference between the tensile stress at the punch top and the compressive stress at the flange at bottom dead center of press forming, as shown in Fig. 2(a). Figure 2(b) shows a schematic diagram of the relationship between the stress and the strain at the punch top and the flange when mass production upper and lower limit materials (highest and lowest strength steels of one grade) are formed by the conventional forming method. Dimensional scattering occurs due to the stress fluctuation at the bottom dead center of press forming when material strength scattering occurs, and this dimensional scattering becomes large as material strength scattering increases. Therefore, reducing the stress on the punch top and flange that causes camber back is effective for reducing the dimensional scattering of camber back.
Camber back of curved part.
Mechanisms of camber back and its scattering; (a) Stress distribution at bottom dead point. (b) Schematic stress-strain curves of lowest and highest strength steel sheets of one grade.
Focusing on the Bauschinger effect, we developed a press forming method that suppresses dimensional scattering, as shown schematically in Fig. 3. The newly-developed method consists of two processes: In the 1st process, the longitudinal radius of curvature is made smaller than that in the 2nd process, and the 2nd process is a restrike forming step. The developed method is a technology for designing the 1st process in such a way that the stresses of the punch top and the flange are reversed at the bottom dead center of the 2nd process, as shown in Fig. 4(a). Figure 4(b) shows the relationship between the stress and the strain at the punch top and the flange when using the upper and lower limit materials. The material strength sensitivity is reduced by inverting the stress in the 2nd process and designing the shape in the 1st process so that the stress generated in each material is the same when forming materials having different strengths. The absolute value of the stress that occurs at the punch top and the flange is smaller than in conventional one-step forming because the Bauschinger effect is expressed in the 2nd process. As a result, the amount of springback is reduced. Based on the above mechanism, it is considered that the developed method can reduce the amount of springback and the material strength sensitivity of springback.
Newly developed 2-step press forming method; (a) 1st process. (b) 2nd process.
Mechanism of camber-back suppression; (a) Stress distribution at bottom dead point of 2nd process. (b) Schematic stress-strain curves of lowest and highest strength steel sheets of one grade.
In order to confirm the effects described in the previous chapter, a press forming analysis was conducted. The finite element software LS-DYNA ver. 9.71 was used for the FEM analysis, and the Yoshida-Uemori model (YU model), which is capable of expressing the Bauschinger effect, was used as the material model. A shell element was used for the FEM analysis. A dynamic explicit method was used for the press forming analysis, and a static implicit method was used for the springback analysis. The longitudinal curved model with a hat cross section shown in Fig. 5 was used. The length in the longitudinal direction was 400 mm, and the standard radius of curvature was 1600 mm at the punch top and 1560 mm at the flange with constant curvature. The cold-rolled steel sheets (thickness of 1.4 mm) with mechanical properties of 590 MPa to 1180 MPa steels shown in Table 1 were used to evaluate the sensitivity of material strength scattering. The press forming process comprised the two processes shown in Fig. 6. The 1st process was drawing, and the 2nd process was form forming with a pad. The Blank Holder Force (BHF) in the 1st process was 490 kN, and the pad load in the 2nd process was 98 kN. In this research, the radius of curvature of the punch top in the 1st process and 2nd process was changed from 1000 mm to 1600 mm with a constant vertical wall height. The amount of camber back was evaluated by the radius of curvature of the punch top.
Model shape and its dimensions; (a) Whole view. (b) Cross section AA′. (c) Side view.
Tool configuration of developed 2-step press forming; (a) 1st process. (b) 2nd process.
Figure 7 shows the tension-compression test results for each material in Table 1 and the stress-strain diagram of the YU model created from the test results. The tension-compression behavior of each material is reproduced by the YU model. In addition, it can be confirmed that the Bauschinger effect becomes large as the material strength increases.
Tension and compression test results for each material; (a) JSC590Y. (b) JSC980Y. (c) JSC1180Y.
Figure 8 shows a comparison of stress-strain curves of the 590 MPa and 1180 MPa steels calculated by the YU model. A large stress difference occurs in the tensile region. On the other hand, the Bauschinger effect is expressed when the stress shifts from tension to compression, and the intersection point of the lines of the 590 MPa and 1180 MPa steels appears. Around this intersection, it is considered that the stress difference between each material is reduced, and as a result, the dimensional scattering between each material is also reduced.
Results for 590 MPa steel and 1180 MPa steel.
Dimensional scattering was examined when the punch top radius of curvature of the 1st process was 1000 mm to 1600 mm and the 2nd process was restrike with the standard 1600 mm. Figure 9 shows the relationship between the punch top radius of curvature of the 1st process and the radius of curvature of the web of the panel after springback formed in the 1st process. For all materials, the radius of curvature after springback was smaller than the radius of curvature 1600 mm of the 2nd process die under the condition that the punch top radius of curvature of the 1st process was 1200 mm or less. Figure 10 shows the relationship between tensile strength and the radius of curvature of the web of the panel after springback formed in the 2nd process for each punch top radius of the 1st process. The dimensional scattering of the 590 MPa to 1180 MPa steels was minimized under the condition that the punch top radius of curvature of the 1st process is 1100 mm. From the above, it was found that dimensional scattering is reduced by forming the longitudinal radius of curvature in the 1st process with a radius of curvature smaller than that of the 2nd process and performing a restrike in the 2nd process.
Effects of punch top radius of curvature of 1st process on radius of curvature of web of panel formed in 1st process.
Effects of tensile strength of steel sheet on radius of curvature of web of panel formed in 2nd process.
In order to investigate the influence of the Bauschinger effect, the conventional method and the developed method were compared using the mold shapes shown in Table 2. The A condition was one-step forming with the standard radius of 1600 mm. The B condition was forming with the standard radius of 1600 mm in the 1st process and 1400 mm, which was smaller than the standard one, in the 2nd process. This condition assumes the conventional estimate process. The C condition was forming with 1100 mm, which is smaller than the standard one, in the 1st process, and the standard radius of 1600 mm in the 2nd process. This condition is the newly-developed forming condition.
Figure 11 shows the relationship between tensile strength and the radius of curvature of the web of the panel after springback formed in the 2nd process for each condition. In the case of the A condition, the radius of curvature becomes large because springback occurs from the standard radius of 1600 mm. In addition, the radius of curvature becomes larger as tensile strength increases. In the case of the B condition, the radius of curvature is smaller than that of the standard radius of 1600 mm because the 2nd process used 1400 mm, which is smaller than the standard one, and dimensional scattering also occurs. The C condition has a smaller radius of curvature than the standard radius of 1600 mm for each tensile strength. However, dimensional scattering was reduced in comparison with both the A condition and the B condition. Under the B condition, the difference in the radius of curvature between the 590 MPa steel and the 1180 MPa steel was 75 mm. In contrast to this, under the C condition, the difference in the radius of curvature between the 590 MPa steel and the 1180 MPa steel was 15 mm. Thus, in comparison with the conventional method, dimensional scattering was reduced by 80% by the developed method.
Amount of camber back in developed forming method and conventional method.
In order to consider the influence of the Bauschinger effect, Fig. 12 shows the FEM analysis results using an isotropic hardening model (IH model) that does not consider the Bauschinger effect for the C condition (developed method). In the case of YU model, the difference in the radius of curvature between the 590 MPa and 1180 MPa steels was 15 mm, but with the IH model, the difference was 37 mm, showing that dimensional scattering was reduced by considering the Bauschinger effect.
Amount of camber back in YU model and IH model.
Figures 13 and 14 show stress distributions at the bottom dead center of the 2nd process for the 590 MPa and 1180 MPa steels under the B condition and C condition. The stress distribution indicates the X-direction thickness average stress in the XYZ space-fixed coordinate system, and the numbers in the figure indicate the average stress of the punch top and flange. The camber back that occurred in this model is the springback that occurred as a structure. Therefore, the stress difference in the thickness direction at the punch top and flange can be ignored. In the conventional method, tensile stress occurred at the punch top and compressive stress occurred at the flange, but in contrast, compressive stress occurred at the punch top and tensile stress occurred at the flange in the developed method.
Stress distributions at bottom dead center in conventional forming method; (a) 590 MPa steel. (b) 1180 MPa steel.
Stress distributions at bottom dead center in developed forming method; (a) 590 MPa steel. (b) 1180 MPa steel.
In order to clarify the mechanism of dimensional scattering reduction under the C condition, Fig. 15 shows the transition of the average stress and average strain at the punch top during press forming under the A condition and the C condition. In the A condition, tensile stress occurred during press forming and reached its maximum at the bottom dead center of press forming. Furthermore, the stress difference between the 590 MPa and 1180 MPa steels at the bottom dead center of press forming was about 360 MPa.
Stress and strain transitions at punch top; (a) Mold shape A. (b) Mold shape C.
On the other hand, in the C condition, the stress reverses and shifts to the compression in the 2nd process. As a result of this stress reversal, the stress difference between the 590 MPa and 1180 MPa steels is reduced to about 120 MPa by the Bauschinger effect. Therefore, under the C condition, the Bauschinger effect occurred in the 2nd process due to the decrease in the radius of curvature in the 1st process, and the stress difference between the materials was reduced. As the result, dimensional scattering was also reduced.
(2) Influence of wall openingIn hat cross section parts curved in the longitudinal direction, camber back and wall opening occur in conjunction with each other. Therefore, we examined the effect of wall opening on camber back. First, the amount of wall opening was measured at a position 30 mm from the punch top in the longitudinal center cross section, as shown in Fig. 16. Figure 17 shows the effect of the punch top radius of curvature of the 1st process and the steel grade on the amount of wall opening after springback of the 2nd process. The wall opening amount increased as the material strength increased.
Definition of wall opening.
Effects of punch top radius of curvature of 1st process and steel grade on wall opening.
In order to suppress the wall opening caused by angular change at the punch shoulder, C-chamfering17) was applied to the punch shoulder (Fig. 18), and the developed method was examined. Figure 19 shows the effect of the punch top radius of curvature of the 1st process and the steel grade on the amount of wall opening after springback of the 2nd process. At each steel grade and punch top radius of curvature in the 1st process, wall opening was significantly suppressed compared with before applying the wall opening countermeasure (C-chamfering) shown in Fig. 17.
Punch R shapes in each forming processes; (a) 1st process. (b) 2nd process.
Effects of punch top radius of curvature of 1st process and steel grade on wall opening.
Next, Fig. 20 shows the effect of the punch top radius of curvature of the 1st process and tensile strength on the radius of curvature of the web of the panel after springback of the 2nd process when the wall opening countermeasure is used. The radii of curvature of the 980 MPa and 1180 MPa steels were larger than in Fig. 10 before the opening countermeasure, and dimensional scattering was suppressed most effectively when the punch top radius of curvature of the 1st process was 1000 mm. From the above, wall opening and camber back occur in conjunction with each other, and when wall opening is suppressed, the radius of curvature becomes large. This indicates that, it is necessary to design the shape of the 1st process in consideration of the influence of wall opening when this method is applied.
Effects of tensile strength on radius of curvature of web of panel formed in 2nd process.
A press experiment was conducted to verify the effect of the developed method described in Chapter 3. Figure 21 shows an example of a press formed product. In order to eliminate the influence of wall opening, the C chamfer shown in Fig. 18 was applied. The materials used in the experiment were the three steels of 590 MPa, 980 MPa and 1180 MPa grade (thickness of 1.4 mm) with the mechanical properties shown in Table 3. Table 4 shows the experimental conditions. Mold shape A indicates the conventional method (without camber back countermeasure), and mold shape B is the developed method.
Curved panel formed in press trial.
Figure 22 shows the effects of each mold shape and tensile strength on the radius of curvature of the web of the panel after springback in the 2nd process. In both the FEM and experimental results, dimensional scattering was reduced by setting the punch top radius of curvature of the 1st process to 1000 mm. In the experiment, the difference in the radius of curvature between the 590 MPa and 1180 MPa steels was reduced by about 95%, from 320 mm to 10 mm. These results verified the effectiveness of the developed method in reducing dimensional scattering.
Effects of tensile strength of steel sheet on radius of curvature of web of panel formed in 2nd processes in experiment and FEM.
A press forming method that utilizes the Bauschinger effect was developed for suppression of the camber back that occurs when forming a part curved in the longitudinal direction. The following conclusions were obtained.
